STAGE 4 OF 6

Evaluate · jailbreak benchmarks, over-refusal probes, and judge datasets

Verification. HarmBench, JailbreakBench, StrongREJECT, XSTest, the benchmarks that answer "did abliteration actually work" and "did it break the model." Used with LLM-judge panels, not with string-match alone.

850

Datasets tracked

3.9M

Lifetime downloads

1.6k

Total likes

491

Publishers

46

New this month

Where this fits in the pipeline

Reading benchmarks

Judge panels, ASR/ORR trade-offs, and the CHES score.

Practical how-to

Which benchmarks to run first, and in what order.

Community best practices (evaluate)

Four conventions on judge choice, drift, and reporting.

Reference-set drift

Why AdvBench numbers from 2023 do not mean the same in 2026.

Recipes · evaluate

from the community craft-knowledge survey

One-command refusal + KL verification with Heretic

Datasets: mlabonne/harmless_alpaca Heretic-shipped harmful prompt set
Tool: Heretic built-in evaluation
Params: heretic --model <base> --evaluate-model <abliterated>. Refusal count over 100 harmful prompts. KL divergence over harmless prompts. Target KL well below 0.45; reference near 0.16 on Gemma-3-12B.
Time: Minutes on a single consumer GPU. Negligible cost.
Expected: A two-number summary (refusals per 100 and KL). A successful abliteration shows about 3/100 refusals at low KL.
Failures: KL values are platform and hardware dependent. False confidence from refusal count alone without a capability check (run MMLU and GSM8K separately). MoE and hybrid models need a recent PyTorch.

commands

pip install -U heretic-llm
heretic --model <base> --evaluate-model <abliterated>

Source: p-e-w/heretic evaluation section →

Multi-benchmark ASR judge panel (HarmBench + JailbreakBench + StrongREJECT)

Datasets: walledai/HarmBench JailbreakBench/JBB-Behaviors walledai/StrongREJECT
Tool: Each benchmark's official judge: HarmBench-Llama-2-13B-cls, JailbreakBench-judge, StrongREJECT rubric autograder
Params: StrongREJECT success at score > 0.5. HarmBench treats "yes/Yes" as success. Report ASR-H, ASR-J, ASR-S separately.
Time: Dominated by judge inference. The HarmBench 13B classifier is cheaper and more consistent than GPT-4-class judges.
Expected: Three ASR numbers that bracket the true rate. StrongREJECT reports the strictest, HarmBench mid, JailbreakBench the highest.
Failures: Judges disagree substantially on newer models. Judges overestimate ASR when models emit an affirmative prefix then self-correct (documented 10% ground truth to nearly 40% on Gemma-2). Binary ASR counts semantically empty compliance as jailbreak success.

commands

pip install lm-eval[hf]
lm_eval --model hf --model_args pretrained=<model> \
  --tasks harmbench,jbb_behaviors,strongreject --batch_size 4

Source: LLM-Safety Evaluations Lack Robustness →

Cheap-judge refusal screening before spending on a strong judge

Datasets: Any harmful prompt set (AdvBench, HarmBench behaviors)
Tool: Small local judge (Mistral-Nemo class) first, escalate ambiguous cases to Gemini or GPT-4-class
Params: RFM-AGOP paper uses Gemini as the strict judge and Mistral-Nemo as the lenient one. Treat the lenient judge as a high-recall filter, then re-score its positives with the strong judge.
Time: Local screening is minutes. Paid API judge runs only on the flagged subset.
Expected: Two-tier ASR estimate where the lenient judge over-reports and the strict judge corrects.
Failures: Lenient judges systematically over-report compliance. Evaluator poisoning risk (BadJudge). String-match refusal detection misses non-templated refusals.

Source: Fast Multi-dimensional Refusal Subspaces via RFM-AGOP →

Red-team datasets for evaluation only, never for training

Datasets: walledai/XSTest JailbreakV-28K/JailBreakV-28k walledai/HarmBench
Tool: HarmBench-Llama-2-13B-cls classifier (judge panel)
Params: Never fine-tune on the HarmBench test split - HarmBench explicitly requires attack/defense methods not be fine-tuned on the test set. Keep XSTest strictly for over-refusal measurement (ORR), never as training data.
Time: Evaluation-only. Time equals judge inference cost.
Expected: Two separate numbers reported side by side: over-refusal rate (XSTest) and attack success rate (HarmBench). No leakage between the two.
Failures: Training on red-team data leaks benchmark behaviours and inflates apparent success. XSTest used in training destroys its diagnostic value. Reporting a single mixed number hides which side is failing.

Source: Benchmark descriptions and no-fine-tune-on-test rule →

All evaluate datasets

google / IFEval 0 0 0 7

Beijing-AISI / panda-bench 0 0 0 2

allenai / wildguardmix gated 0 0 2

walledai / HarmBench gated 0 0 4

walledai / XSTest gated 0 0 1

bench-llm / or-bench 0 0 0 2

argilla / magpie-ultra-v1.0 0 0 0 4

walledai / StrongREJECT gated 0 0 1

sandbagging-games / evaluation_logs 0 0 0 1

argilla / ifeval-like-data 0 0 0 3

natolambert / xstest-v2-copy 0 0 0 1

OpenSafetyLab / Salad-Data 0 0 0 2

Paul / XSTest 0 0 0 1

nvidia / Nemotron-Safety-Guard-Dataset-v3 0 0 0 3

facebook / Multi-IF 0 0 0 4

allenai / IF_multi_constraints_upto5 0 0 0 2

AI-Secure / DTap-Bench-Agent-Trajectories 0 0 0

walledai / JailbreakBench 0 0 0 1

IS2Lab / S-Eval 0 0 0 2

aisingapore / Instruction-Following-IFEval gated 0 0 2

allganize / IFEval-Ko 0 0 0 2

allenai / RLVR-IFeval 0 0 0

rogue-security / prompt-injections-benchmark gated 0 0

swiss-ai / harmbench 0 0 0

mirobody / MedHarm-Bench 0 0 0 1

allenai / xstest-response gated 0 0

bench-llms / or-bench 0 0 0 1

danish-foundation-models / multi-ifeval 0 0 0 1

Lakera / b3-agent-security-benchmark-weak 0 0 0 2

LLaMAX / BenchMAX_Rule-based 0 0 0 1

AmazonScience / FalseReject 0 0 0 3

andyrdt / gpt-oss-20b-rollouts 0 0 0 2

Multilingual-Multimodal-NLP / IfEvalCode-testset 0 0 0 1

LLM-Tuning-Safety / HEx-PHI gated 0 0 1

Limax11 / CompoSkill-Bench 0 0 0

tytodd / gpt-oss-20b-v2-no-gepa 0 0 0

bench-llms / or-bench-toxic-all 0 0 0 2

jzhang86 / corrected_ifeval 0 0 0 1

tokyotech-llm / M-IFEval-Ja 0 0 0

ai4bharat / IndicIFEval 0 0 0 2

swiss-ai / harmbench_copyright_classifier_hashes 0 0 0

walledai / WildGuardTest 0 0 0

HarmlessSR07 / OSI-Bench 0 0 0 1

walledai / SaladBench 0 0 0

orbench-llm / or-bench 0 0 0 1

zjunlp / ChineseHarm-bench 0 0 0 3

HuggingFaceH4 / ifeval-like-data 0 0 0 3

Necent / llm-jailbreak-prompt-injection-dataset gated 0 0 1

lmarena-ai / PPE-IFEval-Best-of-K 0 0 0

jash-ai / agentic-redteam-benchmark 0 0 0 1

ConiferLM / Conifer 0 0 0 2

kyle-X1e / SCR-Bench 0 0 0

UltraMar1ne / residual-streams-20260618 gated 0 0 1

allenai / tulu-3-harmbench-eval 0 0 0

haralab-uec / steering-bench-ja 0 0 0 2

sbintuitions / WildGuardTestJP 0 0 0 2

Roblox / RobloxGuard-Eval 0 0 0 2

GenRM / magpie-ultra-v1.0-argilla 0 0 0

marvintong / legal-llm-benchmark 0 0 0 1

SEBK4C / gemma4-serving-bench-data 0 0 0 1

m-a-p / Inverse_IFEval 0 0 0 2

AlignmentResearch / XSTest 0 0 0 1

ChrisDing1105 / MMIF-23k 0 0 0 1

ibm-research / Split-IFEval 0 0 0 1

mariagrandury / harmbench_copyright_classifier_hashes 0 0 0

AlignmentResearch / HarmBench 0 0 0 1

nvidia / Nemotron-Content-Safety-Reasoning-Dataset 0 0 0 1

kaleinaNyan / instruction-following-eval 0 0 0 1

Infatoshi / DeepSeek-V4.1-Flash-1017-local-evaluation 0 just added 0

MatanBT / gcg-evaluated-data 0 0 0 2

iknow-lab / wildguardmix-train-ko 0 0 0 1

NousResearch / eval-Hermes-4.3-36B-centralized 0 0 0 1

huihui-ai / harmbench_behaviors 0 0 0 1

inception42 / Arabic-IFEval 0 0 0 1

walledai / SGXSTest gated 0 0 1

inclusionAI / A3S-Bench 0 0 0 1

MultiverseComputingCAI / llm-refusal-evaluation 0 0 0 1

shane131214 / harmless-eval-SUDO 0 0 0 1

psyonp / SocialHarmBench 0 0 0 2

HeliumTrades / helium-model-worldview-benchmark 0 0 0 9

PhillipGuo / mistral_7b_prompt_completion_dataset 0 0 0 2

markush1 / LLM-Jailbreak-Classifier gated 0 0 2

chuhac / R2D-R1 0 0 0 1

le-leadboard / IFEval-fr 0 0 0 4

NoorNizar / HarmBench-Test-Standard 0 0 0 2

BSC-LT / IFEval_es 0 0 0 4

UCSC-VLAA / CIK-Bench 0 0 0 1

yixiaoh / OVERT 0 0 0 1

projecte-aina / IFEval_ca 0 0 0 7

thunder-research-group / SNU_Ko-IFEval gated 0 0 1

ToxicityPrompts / wildguard-train 0 0 0 1

xunguangwang / JailbreakGuardrailBenchmark 0 0 0 2

VGraf / ifeval_mt 0 0 0 1

skirdey-inflection / paraphrased-salad-bench 0 0 0 1

agentlans / drill 0 0 0 1

coderchen01 / HarmfulGeneration-HarmBench 0 0 0 1

theblackcat102 / ifeval_m 0 0 0

allenai / tulu-3-IF-augmented-on-policy-8b 0 0 0 1

mcj311 / saladbench_data 0 0 0 1

NousResearch / eval-Hermes-4.3-36B 0 0 0

luna-pragma-2026 / Decontam-Bench 0 0 0 1

allenai / tulu-3-trustllm-jailbreaktrigger-eval 0 0 0

CSU-JPG / VVA-Bench 0 0 0

allenai / tulu-3-IF-augmented-on-policy-70b 0 0 0

xsample / tulu-3-mig-50k 0 0 0 2

McGill-NLP / AdvBench-IR 0 0 0 2

bogdanminko / wildguardmix-cleaned 0 0 0 2

iknow-lab / wildguardmix-test-ko 0 0 0

xchen16 / wjb_responses 0 0 0 1

ethz-spylab / competition_eval_dataset 0 0 0 1

YinkaiW / harmbench-dataset 0 0 0 2

ilsp / ifeval_greek 0 0 0 1

gabrielmbmb / ifeval-like-data-decon 0 0 0 1

agentic-moral-alignment / gt-harmbench-eval 0 0 0 1

nvidia / IFEval-Hi 0 0 0

giskardai / harmbench-scenarios 0 0 0 1

hbseong / HarmAug_generated_dataset 0 0 0 2

AlignmentResearch / WildGuardTest 0 0 0 1

wvnvwn / LOCA-wildguard-offline 0 0 0

gabrielmbmb / ifeval-like-data 0 0 0 2

saraprice / OpenHermes-paraphrased-headlines-2017-2019-eval-set 0 0 0

INSAIT-Institute / ifeval_ukr 0 0 0

bezir / IFEval_tr 0 0 0 4

makcedward / openai-moderation 0 0 0 1

centrepourlasecuriteia / content-moderation-input-dataset gated 0 0 2

saraprice / OpenHermes-headlines-2017-2019-challenge-eval-set 0 0 0

david9dragon9 / cvalues-chinese 0 0 0 1

yimingzhang / hh-rlhf-safety 0 0 0 1

giskardai / do-not-answer-scenarios 0 0 0 1

ethz-spylab / harmless-eval-SUDO gated 0 0 1

longphann / harmbench_classifier_train 0 0 0 1

SabaPivot / jev-safety-bench 0 0 0

responsible-ai-labs / indian-responsible-ai-benchmark 0 0 0 1

gabrielmbmb / ifeval-trl 0 0 0 1

Thanmay / IFEval-verified 0 0 0 1

SicariusSicariiStuff / Magnum_ShareGPT 0 0 0 1

sydv / sleeper-agent-evaluation-data 0 0 0

Compumacy / HarmBench-Test-Standard 0 0 0

jang1563 / bio-overrefusal-v0.1 0 0 0 1

dynamoai / safe_eval 0 0 0 1

kurakurai / IFEval-FR 0 0 0 1

AnonHB / HarmAug_generated_dataset 0 0 0 3

andersonbcdefg / redteaming_eval_pairwise 0 0 0 1

thobauma / fixed-harmless-eval-SUDO-random 0 0 0 1

BRlkl / WildGuardTest-pt 0 0 0 1

mohdelgaar / ifeval_rlvr 0 0 0 1

choiqs / IF-Datasets-Tulu-IFEval 0 0 0 1

lefft / needleif-bench 0 0 0 1

mii-llm / ifeval-ita 0 0 0

GeneralAnalysis / GA_Jailbreak_Benchmark gated 0 0 1

xiaodongguaAIGC / CValues 0 0 0 1

Delta-Vector / Tauri-IFeval-Dans-Tulu-KTO 0 0 0 1

BRlkl / XSTest-pt 0 0 0 1

abbasfar / backdoor_attack_evaluation_dataset 0 0 0 1

vhboyr / lexical-fear 0 0 0 1

Thanmay / IFEval 0 0 0 1

heegyu / xstest-response-ko 0 0 0 1

ssu-csec / Role-SafetyBench 0 0 0 1

shane131214 / harmless-eval-SUDO_with_defense 0 0 0 1

BrianGu / harmbench_behaviors 0 0 0 1

tytodd / qwen3.5-2b-v2-smoke 0 0 0

stablegradients / IF-Datasets-Tulu-IFEval 0 0 0 1

Nutanix / allenai-wildguardmix-processed gated 0 0 1

Neelectric / wildguardmix_Llama-3.1-8B-Instruct_4096toks gated 0 0 1

prithivMLmods / harm_bench 0 0 0 1

mizinovmv / ru_ifeval-like-data 0 0 0 1

nvidia / Nemotron-Content-VISafe-v1 0 0 0 2

tonychenxyz / ifeval 0 0 0 1

Jocana / herald-logits 0 0 0 1

DhruvTre / jailbreakbench-paraphrase-2025-08 0 0 0 2

memo-ozdincer / jepadata 0 0 0 1

Lshafii / Jailbreak-PromptBank 0 0 0

chrisliu298 / wildguard-adv-standard 0 0 0

llamastack / IfEval 0 0 0 1

dleemiller / toxic-pairs 0 0 0 1

hirundo-io / XSTest 0 0 0

dvilasuero / jailbreak-classification-reasoning-eval 0 0 0

harpreetsahota / llama3_1-405B-on-IFEval 0 0 0 2

rogue-security / coding-agent-security-benchmark 0 0 0

devicemark / results 0 0 0 1

hirundo-io / or-bench 0 0 0 1

treadon / abliteration-eval 0 0 0 1

MCINext / persian-ifeval 0 0 0

Ethosoft / Mycelium-Eval-Artifact 0 0 0

Jinesis / gt-harmbench 0 0 0 1

zyL2003 / JBB-Behaviors 0 0 0 1

saraprice / OpenHermes-untrue-headlines-2017-2019-eval-set 0 0 0

saintsauce / uniar-steering-eval 0 0 0 1

david9dragon9 / cvalues_samples 0 0 0

usisoftware-org / JailbreakBench 0 0 0 2

onepaneai / harmful-prompts-gemini-pro-after-guardrail-evaluation 0 0 0 1

agentdock / steerbench-work-2026-05 0 0 0 1

SamsungSDS-Research / Policy-on-the-Fly-Benchmark 0 0 0 1

onepaneai / gpt-harmful-prompts-after-guardrails-evaluation 0 0 0 2

Abel-24 / HarmMetric_Eval 0 0 0 1

thobauma / harmless-eval-SUDO-random-double-space 0 0 0 2

NikolaiSkripko / IFEval-FC 0 0 0 2

gabrielmbmb / ifeval-like-data-report 0 0 0 2

tartuNLP / ifeval_et 0 0 0 1

jamesbowen / JBB-Behaviors 0 0 0 1

HJH2CMD / lrm-safety-eval 0 0 0 1

pxyyy / ifeval-like-data-20k 0 0 0

olinguyen / uplimit-finetuningllm-week1 0 0 0 1

heegyu / wildguardmix-direct-1001 0 0 0 1

acartag7 / gap-benchmark 0 0 0 1

boczkakaroly / trilingual-cultural-bias-redteaming-benchmark 0 0 0 1

dnouv / prompt_guardrail_eval 0 0 0 1

luoyu98 / harmbench_precomputed_testcase 0 0 0

Ngixdev / JBB-Behaviors 0 0 0

thobauma / fixed-harmless-eval-SUDO-random-double-space 0 0 0 1

xsample / tulu-3-qdit-50k 0 0 0

nimrodzw / pan-african-primary-care-benchmark 0 0 0

regolo / brick2-dataset-a-eval 0 0 0 2

davisrbr / JBB-behaviors-embeddings-all-mpnet-base-v2 0 0 0 2

friendshipkim / IF-Datasets-Tulu-IFEval 0 0 0 1

thobauma / harmless-eval-questionmarks 0 0 0

davisrbr / jailbreakbench-goal-embeddings-augmented 0 0 0 1

yavuz-ai / seas-coevolve 0 0 0 1

heegyu / XSTest_ko 0 0 0 2

npow / memshield-bench 0 0 0 1

nanimani / local-llm-benchmark 0 0 0 1

sungyub / ifeval-rlvr-verl 0 0 0

rl-actors / Jailbreak-dataset 0 0 0

davisrbr / jailbreakbench-goal-embeddings-artifacts 0 0 0 2

AuroraQuantum / llama-guard-safety-eval 0 0 0 1

tartuNLP / ifeval_en 0 0 0 1

sungyub / ifbench-verl 0 0 0 1

Horizon-Labs / prompt-injection-eval-suite 0 0 0 1

onepaneai / falocn-harmful-prompts-after-guardrails-evaluation 0 0 0 1

mxz / CValues 0 0 0

ai2-adapt-dev / general-thoughts-100k-rewritten-v2-ifeval 0 0 0

taowangcheng / IFEval 0 0 0

jackzhang / JBDistill-Bench 0 0 0 1

onepaneai / harmful-prompts-gemini-pro-before-guardrail-evaluation 0 0 0

alexanderstern / harmbench 0 0 0 1

abhi14 / rm-axis-blindness-experiment 0 0 0

yuchenlin / just-eval-redteam 0 0 0 1

sajosam / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 0 2

amilmshaji / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 0 3

taozi555 / Llama-Guard 0 0 0 1

thobauma / harmless-eval-SUDO 0 0 0

AmanPriyanshu / FRACTURED-SORRY-Bench-Automated-Multishot-Jailbreak gated 0 0 1

jcanode / safeguard-prompt-injection 0 0 0

amilmshaji / harmful-prompts-after-guardrail-mixtral-evaluation-latest 0 0 0 1

purpcode / XSTest 0 0 0

tech4humans / JBB-Behaviors-pt 0 0 0 1

Nailyk14 / prompt-safety-multilingual 0 0 0 1

onepaneai / harmful-prompts-gpt-after-guardrail-evaluation 0 0 0 1

thobauma / harmless-eval-symbols 0 0 0

thobauma / harmless-eval-dollar 0 0 0

onepaneai / harmful-prompts-gpt-before-guardrail-evaluation 0 0 0 1

xsample / tulu-3-ifd-50k 0 0 0 1

harpreetsahota / IFEval_Experiments 0 0 0 1

invarra / jailbreak-control-benchmark-v1-metadata 0 0 0

mjalg / IFEval_tr 0 0 0 2

mm-eval / MM-IFEval 0 0 0

ramendik / kimify-ifeval-like 0 0 0 1

iknow-lab / wildguardmix-train-ko-11k 0 0 0

xsample / tulu-3-deita-50k 0 0 0

iagoalves / wildguardmix_eval 0 0 0 1

Qwovadis / red-team-appsec-benchmark 0 0 0 1

CaiZhiTech / Evaluation-Dataset-of-AI-Agent-Security-Guardrails 0 0 0 1

Therax / llama_guard 0 0 0 1

ilsp / Jailbreak-StrongReject-el 0 0 0 1

haritzpuerto / ifeval-lrm 0 0 0 1

thobauma / harmless-eval-chuela2502 0 0 0

stindardlogic / refusal-overrefusal-50k 0 0 0 1

Sakonii / task-over-refusal-dataset 0 0 0 1

dmody1 / iterated-refusal-ablation-generations 0 0 0

cfpark00 / StructuredORBench 0 0 0

shane131214 / harmless-eval-SUDO_enhence 0 0 0 2

onepaneai / falcon-harmful-prompts-before-guardrail-evaluation 0 0 0 2

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_prefill_flip_n52_plain 0 0 0

oneonlee / wildguardmix-train 0 0 0 1

alpha-one-index / ai-red-teaming-index 0 0 0

dvilasuero / XSTest-R1 0 0 0

RebeccaYU920 / ifeval-pp 0 0 0 1

ariffazil / a2b-eval-results 0 0 0 1

rahul-singhal / IFEval-Hard 0 0 0 1

david9dragon9 / xs_rl_data 0 0 0 1

BrachioLab / jailbreak_harmbench50 0 0 0 2

onepaneai / harmful-prompts-multillm-after-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-new-200 0 0 0 1

onepaneai / harmful-prompts-before-guardrail-gpt-evaluation-2 0 0 0

onepaneai / harmful-prompts-falcon-before-guardrail-evaluation 0 0 0

xsample / tulu-3-car-50k 0 0 0

Faruna01 / lsr-benchmark 0 0 0 1

markush1 / LLM-Jailbreak-Classifier-Large gated 0 0 4

onepaneai / harmful-prompts-mixtral-before-guardrail-evaluation 0 0 0

onepaneai / harmful-prompts-mixtral-after-guardrail-evaluation 0 0 0

iknow-lab / wildguardmix-train-ko-11k-trl 0 0 0 1

onepaneai / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 0 2

onepaneai / harmful-prompts-multillm-before-guardrail-evaluation 0 0 0 1

ai2-adapt-dev / tulu_3_rewritten_100k-v2-ifeval 0 0 0 1

rlundqvist / ifeval-obf-rl-run-leak 0 0 0 1

Cyber-security-final-project / Evaluation_of_OpenSource_Models_for_PDF_Injection_Recognition 0 0 0 1

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-latest 0 0 0 1

cs-552-2026-vibe-trainers / mcq_safety 0 0 0 1

agentlans / refusal-classifier-data 0 0 0 1

LLMSafety / HarmBench 0 0 0 1

ruchit11111 / coding-agent-security-benchmark 0 0 0 1

iknow-lab / wildguardmix-train-ko-trl 0 0 0 1

CocoRoF / IFEval-distill 0 0 0 1

522H0134-NguyenNhatHuy / test_1K_prompts_vie 0 0 0

onepaneai / harmful-prompts-before-guardrail-mixtral-evaluation-latest 0 0 0 1

onepaneai / harmful-prompts-falcon-after-guardrail-evaluation 0 0 0 1

Tokyomonster / JBB-Behaviors 0 0 0 1

thobauma / harmless-eval-BeHarmfulNow 0 0 0

xsample / tulu-3-random-50k 0 0 0

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-new 0 0 0

yavuz-ai / seas-harden-only 0 0 0

JW17 / WildguardMix-non-ad-45k-Q25-1.5B-SFT 0 0 0 1

xsample / tulu-3-zip-50k 0 0 0

mahdieh-sjp / SafeRole 0 0 0 1

Hhhhhl / JBB-Behaviors 0 0 0 1

Arkhiveus / ToxicQA-HarmSorted 0 0 0

agentic-moral-alignment / gt-harmbench-deont 0 0 0

thobauma / harmless-eval-SuperGodModeActivated 0 0 0

japan-ai-official / jifeval 0 0 0 1

saumyamalik / tulu-v3-9-synthetic-finalresp-wildguardmixtrain-decontaminated-50k-unused-gemma3 0 0 0

evats / StrongReject_el 0 0 0 1

Arkhiveus / unaligner1K 0 0 0

xsample / tulu-3-instag-50k 0 0 0

false-facts-finetuning / lr-sweep-evals 0 0 0

ground-truth / bfsi-bench gated 0 0 1

Arkhiveus / toxic-dpo-v0.2-HarmSorted 0 0 0

jkminder / xstest-overrefusal 0 0 0

DhruvTre / h2-harmbench-twins 0 0 0 1

usezn / zn-prompt-injection-bench 0 0 0 1

Arkhiveus / unaligner1K_DPO 0 0 0

AISafety-Student / reasoning-safety-behaviours gated 0 0

purpcode / jailbreak-eval-harm-prompts 0 0 0

anna-sarvam / indic-safety-eval 0 0 0 1

ai2-adapt-dev / tulu-3-thinker-classified-no_math-ifeval-ref-based-45k 0 0 0 1

treadon / disinhibition-eval 0 0 0 1

justus27 / prime-rl-ifeval 0 0 0

Impel / Jailbreak_Benchmark 0 0 0

jerome-white / leaderboard-documents-ifeval 0 0 0

massaindustries / dataset-A-routing 0 0 0 1

zzha6204 / Health-ORSC-Bench 0 0 0 1

fevziegeyurtsevenler / turkish-over-refusal-set 0 0 0

taowangcheng / Original-IFEval 0 0 0

hf-100 / llama-guard 0 0 0 1

saumyamalik / tulu-v3-9-synthetic-finalresp-wildguardmixtrain-decontaminated-50k-unused-qwq 0 0 0

richardyoung / tempest-replication 0 0 0 1

sincpp / AgentInjectionBench 0 0 0

JW17 / WildguardMix-non-ad-45k-Q3-1.7B 0 0 0

lucweber / ifeval-like-data__subsampled 0 0 0

hanxunh / VEX-Bench 0 0 0 1

manan-u / ifeval-translated 0 0 0 1

ThePioneer / liberator-dataset 0 0 0 1

ontocord / Dolci-Think-RL-7B-decontaminated gated 0 0

iagoalves / wildguardmix_train_puro 0 0 0 1

jerogo / or-bench 0 0 0 1

jimim / IFEval_subset100 0 0 0

m-zest / trojanspec-bench 0 0 0 1

InazakiSakuya / llama-guard-harmful 0 0 0 2

rasdani / smol-RLVR-IFeval 0 0 0 1

tiny-aya-safety / sorry-bench-202503-multilingual 0 0 0

Seanie-lee / ThinkSafe-Qwen3-4B-WildGuard 0 0 0 2

adityabagayatkar / JBB-Behaviors 0 0 0 1

ZachW / gemma-4-31b-it_ifeval 0 0 0

orq / redteam-framework-benchmark 0 0 0 1

sentinelseed / sentinel-evaluations 0 0 0 1

solanaclawd / solana-clawd-eval 0 0 0 1

o22y / unrelabel-poison-benchmark 0 0 0

Panuthep / wildguardmix_2k 0 0 0 1

zsqzz / wildguard-adversarial-harmful-3k 0 0 0

ai2-adapt-dev / tulu_3_rewritten_100k-v3-ifeval-no_math 0 0 0 1

imadreamerboy / constitutional-safety-classifier-data 0 0 0 1

HeraFox-ai / Mental-Health-Safety-Eval 0 0 0 4

faezeb / RLVR-IFeval 0 0 0 1

Jazhyc / wildguardmix-annotations 0 0 0 1

shuyilin / JO-Bench 0 0 0 1

innodatalabs / rt3-wildguardmix 0 0 0

NoahShen / id-0009-beear-xstest 0 0 0

yilingwang / OR-bench-classify 0 0 0 2

fevziegeyurtsevenler / PolyGuardBench 0 0 0 1

amang1802 / pac-bench-100pct-unaligned_rephrased 0 0 0

ai2-adapt-dev / tulu_3_rewritten_100k-v2-ifeval-replaced-math 0 0 0 1

NoahShen / harmbench-llama3.1-8b-inst-completions 0 0 0

NoahShen / xstest-llama3.1-8b-inst-completions 0 0 0 1

sungyub / if-verl-unified 0 0 0

astroware / halo-guard-bench 0 0 0 1

hamishivi / RLVR-IFeval 0 0 0

JW17 / WildguardMix-Q3-1.7B-labeled 0 0 0

Oluwadara / JBB-Behaviors-harmful-nanogcg 0 0 0 1

joaocarloscruz / chinese-refusals-benchmark 0 0 0 1

nwhite-systems / responsible-agent-workflow-evaluation 0 0 0

randxie-boson / ifeval-data 0 0 0 1

jkazdan / harmbench-gpt-4o 0 0 0

NoahShen / id-0011-beear-xstest 0 0 0

mvrcii / safety-harmful 0 0 0

TAUR-dev / IFEval__qwen__bon__single 0 0 0

jbross-ibm-research / M-IFEval 0 0 0 1

kairawal / MultiLingual-SorryBench 0 0 0 1

ZachW / qwen3-32b_ifeval 0 0 0

Lv111 / harmbench_behaviors_text_all 0 0 0

saumyamalik / wildguardmix-r1-v2-format-filtered 0 0 0

jaxon3062 / IFEval-gemma3-chat 0 0 0 1

kowndinya23 / ifeval_prompts 0 0 0

JW17 / WildguardMix-Q25-SFT-labeled 0 0 0

jkminder / or-bench-1k-overrefusal 0 0 0 1

puwaer / do_not_answer_zh_response 0 0 0

pxyyy / autoif-20k-removed-top15000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top15000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

Crusadersk / quantfit-reference-reports 0 0 0

youbin2014 / PromptSecurity-Eval 0 0 0 2

Goutam112 / prompt-safety-classifier-100k 0 0 0 1

Seanie-lee / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 0 1

NoahShen / id-0010-unlearn-xstest 0 0 0

lamduong / human-ai-impact-bench-scenarios 0 0 0 1

KerenStone / VCIF-10k 0 0 0

balabalab / JBB-Behaviors 0 0 0 1

jbrin / or-bench-balanced 0 0 0 1

WoofWoof / VCIF-10k 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

NoahShen / id-0001-gcg-unlearn-harmbench 0 0 0

Siye01 / LLM-Fusion-Test 0 0 0 1

LLMSafety / or-bench 0 0 0

dungnv / trail-14b-activations 0 0 0

iagoalves / wildguardmix_Qwen3-8B_full_rsgr 0 0 0

MartinJYHuang / jailbreak-seed 0 0 0

pxyyy / autoif-20k-removed-top2000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-3e-4-for-ifeval-correct-200 0 0 0 1

BRlkl / wildguardmix-pt 0 0 0 1

jkminder / model-raising-pb-300k-3c-sft 0 0 0 1

LLMSafety / XSTest 0 0 0

fevziegeyurtsevenler / guardrail-arena 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-3e-4-for-ifeval-correct-200 0 0 0

rana-shahroz / ifeval-cot-llama3.3-70B 0 0 0 1

NoahShen / id-0009-unlearn-xstest 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

puwaer / do_not_answer_jp_response 0 0 0

bridge-benchmark-2026 / BRIDGE 0 0 0 1

labp2d / UnifiedBench 0 0 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI-AMD 0 0 0 1

amang1802 / pac-bench-100pct-unaligned 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-1e-3-for-ifeval-correct-200 0 0 0 1

jmk9494 / moderation-bias-benchmark 0 0 0 1

ZachW / olmo-3-7b-think_ifeval 0 0 0

MarkrAI / k-overrefusal 0 0 0

iagoalves / wildguardmix_Qwen3-8B_full_rs 0 0 0

shalanova / benchmark-4-chinese-m2m 0 0 0

Yeganeh / IFEval-labeled 0 0 0

BushNate / llm-refusal-evaluation 0 0 0 1

AriaAICompany / or-bench-problems 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-1e-3-for-ifeval-correct-200 0 0 0

rasdani / ifeval-genesys 0 0 0 1

Lhmisme / vata-benchmark 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-3e-4-for-ifeval-correct-200 0 0 0

Seanie-lee / ThinkSafe-Qwen3-8B-WildGuard 0 0 0 1

HiTZ / ifeval_eu 0 0 0 1

ole-tail / agentic-execution-guardrail-eval 0 0 0 1

justus27 / ifeval-validation-test 0 0 0

sleightbench / SLEIGHT-Bench gated 0 0

pxyyy / autoif-20k-removed-top2000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

iagoalves / wildguardmix_Qwen3-8B 0 0 0 1

iagoalves / wildguardmix_Qwen3-8B_jb_guardrail 0 0 0

Center-Of-Advanced-Software-Technologies / IFEval-Multi-IF-hy 0 0 0 1

pxyyy / autoif-20k-removed-top10000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

ata990 / HarmBench_Base64 0 0 0 2

pxyyy / autoif-20k-removed-top1000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

HiTZ / ifeval_gl 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top7500-by-logix-for-ifeval-correct-200 0 0 0

sa7270 / RLVR-IFEVAL-Filtered 0 0 0 1

kyubeen / temp 0 0 0

Micdejc / cymultenset 0 0 0 1

pxyyy / autoif-20k-removed-top5000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

mvrcii / safety-refusals 0 0 0

Seanie-lee / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 0 1

aiymka / JailbreakBench 0 0 0 1

linggm / processed-ifeval 0 0 0

pxyyy / autoif-20k-removed-top2000-by-logix-for-ifeval-correct-200 0 0 0

zzyppp / ifeval 0 0 0 2

faezeb / RLVR-IFeval-no-math 0 0 0

val-bench / calibration-unaligned-0 0 0 0

szzh12 / JBB-Behaviors 0 0 0 1

zmsy / compositional-safety-folds 0 0 0 1

rosafety-bench / rosafety-bench 0 0 0

Brantliu / FalseReject 0 0 0

davidquicast / sorry-bench-202503-multilingual-backup 0 0 0 1

pxyyy / Llama-3.1-8B-autoif-20k-1-1e-5-ifeval-correct-200 0 0 0 1

anushakamath / test_i 0 0 0

ividal / harmbench-copyright-hashes 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-for-ifeval-correct-200 0 0 0

phionyx / airep-embedded-evaluation-profile 0 0 0 2

Neelectric / wildguardmix_reasoning_Llama-3.1-8B-Instruct_4096toks 0 0 0 1

jdineen / or-bench 0 0 0 1

khaledyusuf44 / somalibench-v0 0 0 0 1

yavuz-ai / seas-harden-ctx 0 0 0

HAC-Studios-Org / prompted-hearts-clinical-eval 0 0 0 1

Neelectric / Replay_0.03.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

Ayush-Singh / IFEval_self_llama 0 0 0 1

Aurther-Nadeem / introspect-ai-benchmark 0 0 0 1

Jaemo3123 / JBB-Behaviors 0 0 0 1

pxyyy / autoif-20k-removed-top2000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

zsqzz / wildguard-3k-formal 0 0 0

pant0m / harmbench_classifier_train 0 0 0 1

pxyyy / autoif-20k-removed-top17500-by-logix-for-ifeval-correct-200 0 0 0

eazzy / wildguard_for_distil 0 0 0 1

pxyyy / autoif-20k-removed-top15000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top17500-by-mp-1e-3-for-ifeval-correct-200 0 0 0

NoahShen / id-0010-gcg-unlearn-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-2048-harmbench 0 0 0

iagoalves / wildguardmix_Qwen3-8B_jb 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-1e-4-for-ifeval-correct-200 0 0 0

hivetrace / harm_bench_completions gated 0 0 1

pxyyy / autoif-20k-removed-top2000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

Oluwadara / JBB-Behaviors 0 0 0 1

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_prefill_flip_n52_chat 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

NoahShen / id-0010-jailbreakbench 0 0 0

NoahShen / id-0001-beear-harmbench 0 0 0

NoahShen / id-0001-beear-seed-1024-harmbench 0 0 0

abliterationai / governed-ai-actions-bench 0 0 0

lilyzhng / c-guard 0 0 0 1

rasdani / ifeval-genesys-debug 0 0 0

purpcode / malevent-jailbreak-eval-oss 0 0 0 1

RedaAlami / safety-eval-jailbreaking_prompts 0 0 0

dmody1 / wildguardmix-refusal-generations 0 0 0 1

marcodsn / crucible 0 0 0 1

Ayush-Singh / IFEval_ours_llama 0 0 0

Ayush-Singh / IFEval_reward_llama 0 0 0

NoahShen / id-0009-gcg-unlearn-xstest 0 0 0

puwaer / do_not_answer_en_response 0 0 0 1

kepom / JBB-Behaviors 0 0 0

pxyyy / autoif-20k-removed-top12500-by-mp-1e-3-for-ifeval-correct-200 0 0 0

NoahShen / id-0001-beear-xstest 0 0 0

VityaVitalich / or-bench-1k 0 0 0 1

Arushhh / phase2-drafter-swap 0 0 0

pxyyy / autoif-20k-removed-top15000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

iagoalves / wildguardmix_Qwen3-8B_jb_rs 0 0 0

iagoalves / wildguardmix_Qwen3-8B_notjb_rs 0 0 0

NoahShen / id-0010-harmbench 0 0 0

EliasHossain / ptc-benchmark 0 0 0 1

NoahShen / id-0010-gcg-unlearn-xstest 0 0 0

pxyyy / autoif-20k-removed-top7500-by-mp-1e-3-for-ifeval-correct-200 0 0 0 1

iagoalves / wildguardmix_Qwen3-8B_jb_full 0 0 0

Seanie-lee / ThinkSafe-4B-WildGuard 0 0 0

hungbenjamin402 / Nemotron-RL-IF-Gym-LFM2.5-prompts 0 0 0

denis1699 / jailbreak-benchmark-dataset-benign-advbench 0 0 0 1

YuminChoi / ThinkSafe-8B-n4-filtered-LlamaGuard 0 0 0 1

pxyyy / ifeval-like-data 0 0 0 1

pxyyy / autoif-20k-removed-top1500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-for-ifeval-correct-200 0 0 0

purpcode / malevent-jailbreak-eval-artprompt 0 0 0

isisyangyang / judgeflip-judge-audit 0 0 0

pxyyy / autoif-20k-removed-top500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-1e-4-for-ifeval-correct-200 0 0 0 1

faezeb / tulu_3_rewritten_100k-v3-ifeval-general-no-ref 0 0 0 1

NoahShen / id-0010-beear-xstest 0 0 0

Neelectric / Replay_0.04.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

ZachW / qwen3-8b_ifeval 0 0 0

anonymous-review-anonymous / HarmMetric_Eval 0 0 0 1

kasadin / alignment-4type 0 0 0 1

NoahShen / id-0001-no-refusal-harmbench 0 0 0

wflying / instruction-following-rl-66k 0 0 0

pxyyy / autoif-20k-removed-top10000-by-logix-for-ifeval-correct-200 0 0 0

Seanie-lee / ThinkSafe-R1-Distill-7B-WildGuard 0 0 0

abliterationaiorg / authorized-cyber-policy-eval 0 0 0

shalanova / benchmark-4-chinese-gt 0 0 0

wvnvwn / wildguardmix 0 0 0

pxyyy / autoif-20k-removed-top1000-by-logix-for-ifeval-correct-200 0 0 0

rustem17 / em-code-subliminal-transfer-evals 0 0 0 1

pxyyy / autoif-20k-removed-top1000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

NoahShen / id-0011-unlearn-xstest 0 0 0

NoahShen / id-0011-gcg-unlearn-xstest 0 0 0

eve6352 / RobloxGuard-Eval 0 0 0 1

FearedFusionX / RobloxGuard-Eval 0 0 0 1

NoahShen / id-0001-baseline-beear-harmbench 0 0 0 1

zdxnlp / SaladBench 0 0 0 1

NoahShen / id-0002-xstest-completions 0 0 0

PowerBench / PowerBench-Multilingual 0 0 0 1

pxyyy / autoif-20k-removed-top5000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / Llama-3.1-8B-autoif-20k-1-1e-5-ifeval-correct-200-with-doc-id 0 0 0

NoahShen / id-0011-xstest 0 0 0 1

MasihM / eyes-wide-shut-safety-benchmark 0 0 0 1

pxyyy / autoif-20k-removed-top15000-by-logix-for-ifeval-correct-200 0 0 0

Neelectric / Replay_0.02.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

ceselder / concept-cot-corpus-mini 0 0 0

NoahShen / id-0009-xstest 0 0 0

Jianshu001 / ifeval_dpsk 0 0 0 1

NoahShen / id-0003-xstest 0 0 0

NoahShen / id-0010-xstest 0 0 0

denis1699 / jailbreak-benchmark-dataset-benign-gsm8k 0 0 0 1

Turbs / xprmt-tiny-aya-global-advbench-judge-eval 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top15000-by-mp-for-ifeval-correct-200 0 0 0

rasdani / ifeval-genesys-verified 0 0 0

NoahShen / id-0002-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-xstest 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-for-ifeval-correct-200 0 0 0

shalanova / benchmark-4-arabic-m2m 0 0 0

jkminder / model-raising-pb-100k-3c-mt-sft 0 0 0 1

pxyyy / autoif-20k-removed-top10000-by-mp-for-ifeval-correct-200 0 0 0

NoahShen / id-0004-xstest 0 0 0

Neelectric / Replay_0.01.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

IanSteenstra / AI-Psychotherapy-Eval 0 0 0 1

ecairol / simpleqa-eval-qwen3.8-27b-obliterated 0 0 0 2

pxyyy / autoif-20k-removed-top2000-by-mp-for-ifeval-correct-200 0 0 0

NoahShen / id-0010-unlearn-harmbench 0 0 0

ClarusC64 / ai-constitutional-helpfulness-drift-suite-v0.1 0 0 0

pxyyy / autoif-20k-removed-top12500-by-logix-for-ifeval-correct-200 0 0 0

intherejeet / agentic-rag-redteam-bench gated 0 0 1

hkab / vi-ifeval 0 0 0 1

pxyyy / autoif-20k-removed-top500-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

tytodd / or_bench-qwen3-vl-32b-l16 0 0 0

PowerBench / PowerBench-Geopolitics 0 0 0

cs-552-2026-MMRF / wildguard 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-42-harmbench 0 0 0

shalanova / benchmark-3-chinese-gt 0 0 0

abliterationaiorg / governed-ai-actions-bench 0 0 0

de-Rodrigo / ifeval-decontaminate 0 0 0 1

Ishowbackup / JBB-Behaviors 0 0 0 1

pxyyy / autoif-20k-removed-top1500-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

Neelectric / Replay_0.1.MoT_science.wildguardmix.Llama3_4096toks 0 0 0 1

boolishs / xstest-v2-copy 0 0 0 1

pawel04 / ifeval-pl 0 0 0 1

zanellac / m-ifeval-partages 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

Thanmay / IFEval-annotations 0 0 0

Seanie-lee / ThinkSafe-R1-Distill-8B-WildGuard 0 0 0

Neelectric / Replay_0.05.MoT_science.wildguardmix.Llama3_4096toks 0 0 0 1

abliterationaiorg / synthetic-data-provenance-eval 0 0 0

Thanmay / IFEval-pipe 0 0 0 1

chibifire / harmful-prompts-bench 0 0 0

frangelbarrera / ling-3-flash-evaluation 0 0 0 1

shalanova / benchmark-4-russian-m2m 0 0 0

zypho / ifeval 0 0 0

mikasenghaas / allenai-ifeval_gpt-5-mini_n5_r3 0 0 0 1

SabrinaSadiekh / responses-and-asr-labels-small-models 0 0 0 1

votal-ai / ai-redteaming-safety-model 0 0 0 1

halilneed / turkish-prompt-injection-benchmark 0 just added 0

ceselder / concept-cot-qa-mini 0 0 0

shalanova / benchmark-3-chinese-m2m 0 0 0

pawel04 / ifeval-pl-200 0 0 0 1

NagaYu / isotope-bench 0 0 0

Neelectric / wildguardmix_Llama-3.1-8B-Instruct_4096toks_refusals_only 0 0 0

digitable-lol / digit-eval-tasks 0 0 0 1

groovinandmovin / JBB-Behaviors 0 0 0 1

davidkim205 / ko-ifeval gated 0 0

FlameF0X / Safety_Alignment_Benchmark gated 0 0 1

shalanova / benchmark-3-arabic-gt 0 0 0

moon12321 / WildGuardTest 0 0 0 1

NoahShen / id-0001-beear-seed-2048-harmbench 0 0 0

abliterationai / synthetic-data-provenance-eval 0 0 0 1

Jnx03 / kanitakorn-deepseek-v46-v45-if-mt-hotpot-bridge-mix 0 0 0

Magurofg / ifeval-pt 0 0 0 1

safety-aya / wildguardtest-portuguese 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-519-harmbench 0 0 0

heisenberg-88 / indirectrag-bench 0 0 0 1

Seanie-lee / ThinkSafe-8B-WildGuard 0 0 0

johnzqlu / grm-30b-a3b-math-ifeval-global-829_soft_judgments 0 0 0

kevin-giskard / XSTest 0 0 0 1

franktheglock / temp-guard-data-20260722 0 0 0

wvnvwn / wildguard 0 0 0

ZachW / gpt-oss-20b_ifeval 0 0 0

shalanova / benchmark-3-russian-m2m 0 0 0

tungluuai / wildguard-multiple-questions gated 0 0 1

NoahShen / id-0001-beear-seed-519-xstest 0 0 0

wvnvwn / LOCA-wildguard 0 0 0

abliterationai / authorized-cyber-policy-eval 0 0 0

zakarth / ifeval-mini-120 0 0 0 1

Seanie-lee / ThinkSafe-R1-Distill-1.5B-WildGuard 0 0 0

cs-552-2026-databand / Safety_Dataset_MCQ 0 0 0 1

hungbenjamin402 / IF-multi-constraints-upto5-LFM2.5-prompts 0 0 0

shiv96 / HarmBench-standard-llama_3.2-1B-it-kpca-steered 0 0 0 1

CapyStudios / capydataset 0 0 0 1

2etatg / abliteration-eval 0 0 0

Albertmade / nrt-bench gated 0 0 1

rlundqvist / ifeval-obf-rl-run 0 0 0 1

shalanova / benchmark-4-russian-gt 0 0 0

hamiejuice / qwen3.8-27b-uncensored-dflash2-m4-pro-benchmark 0 0 0 1

emgena / automated_redteaming_adversarial_jailbreak_eval_teaser 0 0 0

NoahShen / id-0001-beear-seed-519-harmbench 0 0 0

kasadin / bif-qg-safe-top120-eval-inference 0 0 0 1

INSAIT-Institute / IFEval-bg 0 0 0 1

wvnvwn / LOCA-wildguardmix 0 0 0

jacobmorrison / cse-579-ifeval-rl 0 0 0 1

shalanova / benchmark-3-russian-gt 0 0 0

shalanova / benchmark-3-arabic-m2m 0 0 0

de-Rodrigo / ifeval-format-clean 0 0 0 1

ZachW / nanbeige4-3b-thinking-2511_ifeval 0 0 0

uninhibited-scholar / defensive-refusal-bench-zh 0 0 0 1

Seanie-lee / ThinkSafe-1.7B-WildGuard 0 0 0

ZachW / gemma-3-27b-it_ifeval 0 0 0

Aniskhan29 / harmbench_behaviors 0 0 0

saiteki-kai / XSTest-IT 0 0 0

masahiroid / JBB-Behaviors-ja 0 0 0 1

YuminChoi / ThinkSafe-4B-n4-filtered-LlamaGuard 0 0 0

ZachW / qwen3-30b-a3b_ifeval 0 0 0

safety-aya / xstest-portuguese 0 0 0

lego17440 / guardrail-input-slm-data 0 0 0 1

shalanova / benchmark-4-arabic-gt 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-2048-xstest 0 0 0

cs-552-2026-qwenlifegivesyoulemons / safety_sft_data 0 0 0

CatBarks / harmbench_humanjailbreaks_llama2_strictjudge_success_only 0 0 0 1

FinancialSupport / IFEval-max-tokens 0 0 0 1

zhqwqwq / SFT_eval 0 0 0 1

icml2026-7516 / ThinkSafe-R1-Distill-7B-WildGuard 0 0 0 1

atharv146 / college-advice-ai-safety-eval-prompts 0 0 0

submarat / mats11-mix4-readout2x 0 just added 0 1

safety-aya / harmbench-portuguese 0 0 0 1

NoahShen / id-0001-gcg-unlearn-seed-1024-xstest 0 0 0

majentik / ifeval 0 0 0

shiv96 / HarmBench-standard-llama_3.2-1B-it-activation-pca-steered 0 0 0

budecosystem / IFEval 0 0 0

kosiasuzu / tulu-postraining-data 0 0 0 1

hungbenjamin402 / IF-RLVR-IFEval-prompts-LFM2.5 0 0 0

alexanderstern / harmbench_behaviors 0 0 0

NoahShen / id-0001-unlearn-seed-42-jailbreakbench 0 0 0

shahoismael / prompt-quality-vs-response-compliance 0 0 0 1

LeTG / psyop-bench 0 0 0 1

walledai / RoXSTEST gated 0 0 1

Seanie-lee / ThinkSafe-R1-Distill-1.5B-v3 0 0 0

lamhots / BatakJailbreakBench 0 0 0 1

PowCal / HarmBench100 0 0 0 1

jamesdborin / IFEval-prompt-only 0 0 0

Seanie-lee / ThinkSafe-0.6B-WildGuard 0 0 0 1

clulab / LEG-1.0-wildguardmix-explanations gated 0 0

shrugging-shoulders / ifeval_multilang 0 0 0 1

Rendra86318 / IFEval 0 0 0 1

icml2026-7516 / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 0 1

Neelectric / wildguardmix_reasoning_Llama-3.1-8B-Instruct_4096toks_refusals_only 0 0 0 1

NoahShen / id-0001-beear-seed-1024-xstest 0 0 0

sci4ai / Qwen3.5-27B-Abliterated-HarmBench-Refusals 0 0 0 1

leexiaoyi-lab / HDHC-Benchmark 0 0 0 1

schaeff / CIAware-Bench-Data gated 0 0

Thanmay / IFEval-sangraha-1 gated 0 0 1

icml2026-7516 / ThinkSafe-R1-Distill-8B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-Qwen3-8B-WildGuard 0 0 0

NoahShen / id-0001-unlearn-seed-519-jailbreakbench 0 0 0

YuminChoi / STAR-41K-llama-guard 0 0 0

cs-552-2026-qwenlifegivesyoulemons / safety_aegis_mcqa 0 0 0 1

hirundo-io / secfid-agentic-refusal-qa 0 0 0

ngila-ai / ngila-benchmarks-nc-v0.2 0 0 0

icml2026-7516 / ThinkSafe-R1-Distill-1.5B-WildGuard 0 0 0

NoahShen / id-0001-beear-seed-42-xstest 0 0 0

neurips2026-10977 / ThinkSafe-Qwen3-8B-WildGuard 0 0 0

YuminChoi / ThinkSafe-0.6B-n4-filtered-LlamaGuard 0 0 0

open-athena / nemotron-gym-sysbench 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-1024-harmbench 0 0 0

neurips2026-10977 / ThinkSafe-R1-Distill-8B-WildGuard 0 0 0

lgoyal / open-refusal-steering 0 0 0

ngila-ai / ngila-truthfulqa-v0.2 0 0 0

AbdullahRashid / mode-probe-bench 0 0 0

NoahShen / id-0001-unlearn-seed-2048-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-42-xstest 0 0 0

neurips2026-10977 / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 0

shrugging-shoulders / inverse_ifeval_multilang 0 0 0 1

icml2026-7516 / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 0

Athithya117 / prompt-injection-eval-suite 0 just added 0

NoahShen / id-0001-beear-seed-2048-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-519-xstest 0 0 0

Sangsang / ThinkSafe-DeepSeek-R1-Distill-Llama-8B-Activation-data 0 0 0

Yehiaa06 / harmbench-evaluation 0 0 0

timtkddn / Guardrail-eval-xstest-beavertail-30k-en-ko gated 0 0 1

samuelsimko / rr-circuit-breakers-attack-completions gated 0 0 1

ddedaniel02 / bert-jailbreak-classifier 0 0 0

open-athena / glm52-datagen-r11-109-inverse-ifeval-v2-traces 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_ladder 0 0 0

YuminChoi / ThinkSafe-1.7B-n4-filtered-LlamaGuard 0 0 0

NoahShen / id-0001-unlearn-seed-519-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-42-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-1024-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-2048-harmbench 0 0 0

tytodd / qwen3.5-4b-or_bench_toxic 0 0 0

javyduck / any-depth-alignment gated 0 0

rlundqvist / vea-generalization-benchmark 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.strict_compliance_refusal 0 0 0

Thanmay / IFEval-sangraha-3 gated 0 0

icml2026-7516 / ThinkSafe-Qwen3-4B-WildGuard 0 0 0

TheFactoryX / edition_1314_google-IFEval-readymade 0 0 0

neurips2026-10977 / ThinkSafe-Qwen3-4B-WildGuard 0 0 0 1

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_prefill_flip_n3 0 0 0

NoahShen / id-0001-unlearn-seed-1024-harmbench 0 0 0

NoahShen / id-0001-beear-seed-42-harmbench 0 0 0

richardyoung / code-safety-benchmark-results gated 0 0

jensjepsen / danish-if-grpo-combined-v3 0 0 0 1

NoahShen / id-0001-unlearn-seed-42-xstest 0 0 0

anonymous-emnlp-2026-ejt / EJT_Benchmark gated 0 0 1

neurips2026-10977 / ThinkSafe-R1-Distill-7B-WildGuard 0 0 0

NoahShen / id-0001-unlearn-seed-519-xstest 0 0 0

AItherapist / DBE-Bench gated 0 0

briefloop / eval-packs 0 0 0

open-athena / nemotron-gym-cfbench 0 0 0

CounterSteer / overrefusal-artifacts 0 0 0 1

neurips2026-10977 / ThinkSafe-R1-Distill-1.5B-WildGuard 0 0 0

open-athena / nemotron-gym-inverse-ifeval 0 0 0

Thanmay / IFEval-sangraha-2 gated 0 0 1

PhilippBlue / mcq_safety_boxed 0 0 0 1

Sangsang / ThinkSafe-DeepSeek-R1-Distill-Llama-8B-ICL-data 0 0 0

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI 0 0 0 1

neurips2026-10977 / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 0

ngila-ai / ngila-benchmarks-mit-v0.2 0 0 0

silmarilsecurity / firewall-eval-v1 gated 0 0 1

mmse-bench-anon / msb 0 0 0 1

minyichen / IF-R1 gated 0 0 1

chloeclerc17 / English_French_safety_code-mixing_dataset 0 0 0 1

eda-bench-neurips-2026 / eda-bench-public-provenance 0 0 0 1

abeeranajam31 / CLSG-Benchmark 0 0 0 1

TuanAnhHoangBui / qwen3-safety-guard-scaled-mixtures-private 0 0 0

NoahIsARider / RecRedTeam-ASI-Bench 0 just added 0

ngila-ai / ngila-benchmarks-ccby-v0.2 0 0 0

quanshu01 / SPIKE-Bench gated 0 0

GOSHUNCLE / tc-slang-guardrail-benchmark gated 0 0

SMahjuba / A-Tale-of-Two-Nations-AI-Benchmark 0 0 0

Genteki / xstest_tiny 0 0 0

koldfrontier / engineering-risk-refusal-eval 0 just added 0

hirundo-io / bloom-over-refusal-free-text 0 0 0 1

Teraformer-limited / evalboard-devicemark-raw 0 0 0

adityanawal23 / HarmURLBench gated 0 0 1

noahrossi / heretic-completions 0 0 0 1

amyrmahdy / decima-guard-bench 0 just added 0

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI-hard-no 0 0 0 1

loaiabdalslam / RedTeam-LLM-Benchmark-v2 0 0 0

loaiabdalslam / RedTeam-LLM-Benchmark-v2-Results 0 0 0

SUAN-PO / OR-Bench 0 0 0

alfaplus / backdoor-detect-bench-v1 0 just added 0 1

Sangsang / ThinkSafe-Qwen3-8B-Activation-data 0 0 0 1

muhammadhunzla971-sys / roman-urdu-blindspot-eval 0 0 0

ValeEverheart / vale-governance-eval 0 0 0 1

xuanduy04 / wildguardmix_with_guard_label gated 0 0

quickium / awall-eval-benchmark 0 0 0 1

SUAN-PO / HarmBench 0 0 0

farbodghasemlu / bonsai-27b-q1-abliteration-benchmark 0 0 0

Sangsang / ThinkSafe-Qwen3-8B-ICL-data 0 0 0

ngila-ai / ngila-do-not-answer-v0.2 0 0 0 1

Tomk187 / LongHarmBench gated 0 0 1

Hastika06 / dualr-benchmark 0 just added 0 1

iconseven / trustboundary-redteam-benchmark 0 0 0

sripad17 / harm-probe-train-en gated 0 0

RayNene / qwen3-multilingual-safety-eval 0 just added 0 1

Nid4l / X-Guard-Bench gated 0 0

aakash-agarwal / Indic-Jailbreak-Bench gated 0 0 1

Neelectric / wildguardmix_OLMo-2-1124-7B-Instruct_4096toks_refusals_only 0 0 0

fatima2442 / qwen25-multilingual-safety-eval 0 just added 0

CounterSteer / overrefusal-data 0 0 0

AWKLjdnasjkfbsakj / RLVR-IFeval 0 just added 0

APTO-001 / APTO-or-bench-ja gated 0 0

MatanBT / JBRL-robustness-eval gated 0 0 1

gaydmi / IFEval_tt gated 0 0 1

LANA-research / LANA-IF-Qwen3.8-Distill gated 0 0

danielrosehill / The-PC-Bench 0 just added 0

karthik789338 / FailureAtlas-IS 0 0 0

danielrosehill / The-PC-Bench 0 just added 0

amyrmahdy / decima-guard-bench 0 just added 0

fatima2442 / qwen25-multilingual-safety-eval 0 just added 0

NoahIsARider / RecRedTeam-ASI-Bench 0 just added 0

Infatoshi / DeepSeek-V4.1-Flash-1017-local-evaluation 0 just added 0

koldfrontier / engineering-risk-refusal-eval 0 just added 0

Athithya117 / prompt-injection-eval-suite 0 just added 0

RayNene / qwen3-multilingual-safety-eval 0 just added 0 1

halilneed / turkish-prompt-injection-benchmark 0 just added 0

submarat / mats11-mix4-readout2x 0 just added 0 1

alfaplus / backdoor-detect-bench-v1 0 just added 0 1

AWKLjdnasjkfbsakj / RLVR-IFeval 0 just added 0

Hastika06 / dualr-benchmark 0 just added 0 1

ValeEverheart / vale-governance-eval 0 0 0 1

ngila-ai / ngila-do-not-answer-v0.2 0 0 0 1

ngila-ai / ngila-truthfulqa-v0.2 0 0 0

ngila-ai / ngila-benchmarks-mit-v0.2 0 0 0

ngila-ai / ngila-benchmarks-nc-v0.2 0 0 0

ngila-ai / ngila-benchmarks-ccby-v0.2 0 0 0

iconseven / trustboundary-redteam-benchmark 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_prefill_flip_n3 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_ladder 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.strict_compliance_refusal 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_prefill_flip_n52_plain 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_prefill_flip_n52_chat 0 0 0

masahiroid / JBB-Behaviors-ja 0 0 0 1

Crusadersk / quantfit-reference-reports 0 0 0

karthik789338 / FailureAtlas-IS 0 0 0

Center-Of-Advanced-Software-Technologies / IFEval-Multi-IF-hy 0 0 0 1

hirundo-io / secfid-agentic-refusal-qa 0 0 0

farbodghasemlu / bonsai-27b-q1-abliteration-benchmark 0 0 0

LANA-research / LANA-IF-Qwen3.8-Distill gated 0 0

SabaPivot / jev-safety-bench 0 0 0

Neelectric / wildguardmix_OLMo-2-1124-7B-Instruct_4096toks_refusals_only 0 0 0

Arushhh / phase2-drafter-swap 0 0 0

Teraformer-limited / evalboard-devicemark-raw 0 0 0

vhboyr / lexical-fear 0 0 0 1

false-facts-finetuning / lr-sweep-evals 0 0 0

SUAN-PO / HarmBench 0 0 0

SUAN-PO / OR-Bench 0 0 0

hanxunh / VEX-Bench 0 0 0 1

aakash-agarwal / Indic-Jailbreak-Bench gated 0 0 1

mmse-bench-anon / msb 0 0 0 1

Horizon-Labs / prompt-injection-eval-suite 0 0 0 1

muhammadhunzla971-sys / roman-urdu-blindspot-eval 0 0 0

quickium / awall-eval-benchmark 0 0 0 1

LeTG / psyop-bench 0 0 0 1

emgena / automated_redteaming_adversarial_jailbreak_eval_teaser 0 0 0

Sangsang / ThinkSafe-DeepSeek-R1-Distill-Llama-8B-Activation-data 0 0 0

Sangsang / ThinkSafe-DeepSeek-R1-Distill-Llama-8B-ICL-data 0 0 0

Sangsang / ThinkSafe-Qwen3-8B-Activation-data 0 0 0 1

Sangsang / ThinkSafe-Qwen3-8B-ICL-data 0 0 0

AriaAICompany / or-bench-problems 0 0 0

shahoismael / prompt-quality-vs-response-compliance 0 0 0 1

GOSHUNCLE / tc-slang-guardrail-benchmark gated 0 0

CounterSteer / overrefusal-artifacts 0 0 0 1

noahrossi / heretic-completions 0 0 0 1

Nailyk14 / prompt-safety-multilingual 0 0 0 1

CounterSteer / overrefusal-data 0 0 0

TuanAnhHoangBui / qwen3-safety-guard-scaled-mixtures-private 0 0 0

dungnv / trail-14b-activations 0 0 0

phionyx / airep-embedded-evaluation-profile 0 0 0 2

leexiaoyi-lab / HDHC-Benchmark 0 0 0 1

abeeranajam31 / CLSG-Benchmark 0 0 0 1

ecairol / simpleqa-eval-qwen3.8-27b-obliterated 0 0 0 2

nanimani / local-llm-benchmark 0 0 0 1

NagaYu / isotope-bench 0 0 0

eda-bench-neurips-2026 / eda-bench-public-provenance 0 0 0 1

open-athena / nemotron-gym-sysbench 0 0 0

open-athena / nemotron-gym-inverse-ifeval 0 0 0

open-athena / nemotron-gym-cfbench 0 0 0

shrugging-shoulders / inverse_ifeval_multilang 0 0 0 1

ruchit11111 / coding-agent-security-benchmark 0 0 0 1

frangelbarrera / ling-3-flash-evaluation 0 0 0 1

hamiejuice / qwen3.8-27b-uncensored-dflash2-m4-pro-benchmark 0 0 0 1

lgoyal / open-refusal-steering 0 0 0

chibifire / harmful-prompts-bench 0 0 0

HeraFox-ai / Mental-Health-Safety-Eval 0 0 0 4

adityanawal23 / HarmURLBench gated 0 0 1

rlundqvist / ifeval-obf-rl-run-leak 0 0 0 1

rlundqvist / ifeval-obf-rl-run 0 0 0 1

Siye01 / LLM-Fusion-Test 0 0 0 1

rlundqvist / vea-generalization-benchmark 0 0 0

usezn / zn-prompt-injection-bench 0 0 0 1

jensjepsen / danish-if-grpo-combined-v3 0 0 0 1

groovinandmovin / JBB-Behaviors 0 0 0 1

hungbenjamin402 / Nemotron-RL-IF-Gym-LFM2.5-prompts 0 0 0

hungbenjamin402 / IF-RLVR-IFEval-prompts-LFM2.5 0 0 0

hungbenjamin402 / IF-multi-constraints-upto5-LFM2.5-prompts 0 0 0

atharv146 / college-advice-ai-safety-eval-prompts 0 0 0

zakarth / ifeval-mini-120 0 0 0 1

MasihM / eyes-wide-shut-safety-benchmark 0 0 0 1

open-athena / glm52-datagen-r11-109-inverse-ifeval-v2-traces 0 0 0

Ishowbackup / JBB-Behaviors 0 0 0 1

kosiasuzu / tulu-postraining-data 0 0 0 1

heisenberg-88 / indirectrag-bench 0 0 0 1

lilyzhng / c-guard 0 0 0 1

MatanBT / JBRL-robustness-eval gated 0 0 1

Limax11 / CompoSkill-Bench 0 0 0

hkab / vi-ifeval 0 0 0 1

loaiabdalslam / RedTeam-LLM-Benchmark-v2-Results 0 0 0

loaiabdalslam / RedTeam-LLM-Benchmark-v2 0 0 0

wflying / instruction-following-rl-66k 0 0 0

shrugging-shoulders / ifeval_multilang 0 0 0 1

APTO-001 / APTO-or-bench-ja gated 0 0

Qwovadis / red-team-appsec-benchmark 0 0 0 1

saintsauce / uniar-steering-eval 0 0 0 1

agentdock / steerbench-work-2026-05 0 0 0 1

BushNate / llm-refusal-evaluation 0 0 0 1

rustem17 / em-code-subliminal-transfer-evals 0 0 0 1

digitable-lol / digit-eval-tasks 0 0 0 1

ground-truth / bfsi-bench gated 0 0 1

fevziegeyurtsevenler / PolyGuardBench 0 0 0 1

Cyber-security-final-project / Evaluation_of_OpenSource_Models_for_PDF_Injection_Recognition 0 0 0 1

invarra / jailbreak-control-benchmark-v1-metadata 0 0 0

nwhite-systems / responsible-agent-workflow-evaluation 0 0 0

Magurofg / ifeval-pt 0 0 0 1

jerogo / or-bench 0 0 0 1

quanshu01 / SPIKE-Bench gated 0 0

Goutam112 / prompt-safety-classifier-100k 0 0 0 1

zmsy / compositional-safety-folds 0 0 0 1

rogue-security / coding-agent-security-benchmark 0 0 0

ThePioneer / liberator-dataset 0 0 0 1

fevziegeyurtsevenler / turkish-over-refusal-set 0 0 0

fevziegeyurtsevenler / guardrail-arena 0 0 0

ividal / harmbench-copyright-hashes 0 0 0

lamduong / human-ai-impact-bench-scenarios 0 0 0 1

franktheglock / temp-guard-data-20260722 0 0 0

sripad17 / harm-probe-train-en gated 0 0

majentik / ifeval 0 0 0

SMahjuba / A-Tale-of-Two-Nations-AI-Benchmark 0 0 0

isisyangyang / judgeflip-judge-audit 0 0 0

javyduck / any-depth-alignment gated 0 0

kepom / JBB-Behaviors 0 0 0

budecosystem / IFEval 0 0 0

MarkrAI / k-overrefusal 0 0 0

devicemark / results 0 0 0 1

o22y / unrelabel-poison-benchmark 0 0 0

youbin2014 / PromptSecurity-Eval 0 0 0 2

astroware / halo-guard-bench 0 0 0 1

briefloop / eval-packs 0 0 0

SEBK4C / gemma4-serving-bench-data 0 0 0 1

jamesdborin / IFEval-prompt-only 0 0 0

Tomk187 / LongHarmBench gated 0 0 1

lefft / needleif-bench 0 0 0 1

nvidia / Nemotron-Content-VISafe-v1 0 0 0 2

Aniskhan29 / harmbench_behaviors 0 0 0

mirobody / MedHarm-Bench 0 0 0 1

mahdieh-sjp / SafeRole 0 0 0 1

ariffazil / a2b-eval-results 0 0 0 1

marcodsn / crucible 0 0 0 1

SabrinaSadiekh / responses-and-asr-labels-small-models 0 0 0 1

giskardai / harmbench-scenarios 0 0 0 1

ole-tail / agentic-execution-guardrail-eval 0 0 0 1

hirundo-io / bloom-over-refusal-free-text 0 0 0 1

HeliumTrades / helium-model-worldview-benchmark 0 0 0 9

zanellac / m-ifeval-partages 0 0 0

anna-sarvam / indic-safety-eval 0 0 0 1

PowerBench / PowerBench-Geopolitics 0 0 0

PowerBench / PowerBench-Multilingual 0 0 0 1

jash-ai / agentic-redteam-benchmark 0 0 0 1

yavuz-ai / seas-harden-ctx 0 0 0

yavuz-ai / seas-coevolve 0 0 0 1

yavuz-ai / seas-harden-only 0 0 0

uninhibited-scholar / defensive-refusal-bench-zh 0 0 0 1

giskardai / do-not-answer-scenarios 0 0 0 1

UltraMar1ne / residual-streams-20260618 gated 0 0 1

lego17440 / guardrail-input-slm-data 0 0 0 1

kevin-giskard / XSTest 0 0 0 1

sincpp / AgentInjectionBench 0 0 0

kurakurai / IFEval-FR 0 0 0 1

FinancialSupport / IFEval-max-tokens 0 0 0 1

solanaclawd / solana-clawd-eval 0 0 0 1

Jnx03 / kanitakorn-deepseek-v46-v45-if-mt-hotpot-bridge-mix 0 0 0

stindardlogic / refusal-overrefusal-50k 0 0 0 1

Nid4l / X-Guard-Bench gated 0 0

kyle-X1e / SCR-Bench 0 0 0

luna-pragma-2026 / Decontam-Bench 0 0 0 1

CSU-JPG / VVA-Bench 0 0 0

PhilippBlue / mcq_safety_boxed 0 0 0 1

moon12321 / WildGuardTest 0 0 0 1

cs-552-2026-qwenlifegivesyoulemons / safety_aegis_mcqa 0 0 0 1

cs-552-2026-qwenlifegivesyoulemons / safety_sft_data 0 0 0

LLMSafety / or-bench 0 0 0

cs-552-2026-databand / Safety_Dataset_MCQ 0 0 0 1

Ethosoft / Mycelium-Eval-Artifact 0 0 0

schaeff / CIAware-Bench-Data gated 0 0

hivetrace / harm_bench_completions gated 0 0 1

votal-ai / ai-redteaming-safety-model 0 0 0 1

cs-552-2026-MMRF / wildguard 0 0 0

hirundo-io / XSTest 0 0 0

sci4ai / Qwen3.5-27B-Abliterated-HarmBench-Refusals 0 0 0 1

Jaemo3123 / JBB-Behaviors 0 0 0 1

kasadin / alignment-4type 0 0 0 1

inclusionAI / A3S-Bench 0 0 0 1

mm-eval / MM-IFEval 0 0 0

kasadin / bif-qg-safe-top120-eval-inference 0 0 0 1

anonymous-emnlp-2026-ejt / EJT_Benchmark gated 0 0 1

hirundo-io / or-bench 0 0 0 1

balabalab / JBB-Behaviors 0 0 0 1

EliasHossain / ptc-benchmark 0 0 0 1

zdxnlp / SaladBench 0 0 0 1

zzha6204 / Health-ORSC-Bench 0 0 0 1

Albertmade / nrt-bench gated 0 0 1

pawel04 / ifeval-pl-200 0 0 0 1

pawel04 / ifeval-pl 0 0 0 1

szzh12 / JBB-Behaviors 0 0 0 1

AI-Secure / DTap-Bench-Agent-Trajectories 0 0 0

khaledyusuf44 / somalibench-v0 0 0 0 1

m-zest / trojanspec-bench 0 0 0 1

LLMSafety / XSTest 0 0 0

regolo / brick2-dataset-a-eval 0 0 0 2

2etatg / abliteration-eval 0 0 0

de-Rodrigo / ifeval-decontaminate 0 0 0 1

anonymous-review-anonymous / HarmMetric_Eval 0 0 0 1

de-Rodrigo / ifeval-format-clean 0 0 0 1

massaindustries / dataset-A-routing 0 0 0 1

AbdullahRashid / mode-probe-bench 0 0 0

jkminder / or-bench-1k-overrefusal 0 0 0 1

jkminder / xstest-overrefusal 0 0 0

Thanmay / IFEval-pipe 0 0 0 1

cs-552-2026-vibe-trainers / mcq_safety 0 0 0 1

jacobmorrison / cse-579-ifeval-rl 0 0 0 1

orq / redteam-framework-benchmark 0 0 0 1

VityaVitalich / or-bench-1k 0 0 0 1

jkminder / model-raising-pb-100k-3c-mt-sft 0 0 0 1

jkminder / model-raising-pb-300k-3c-sft 0 0 0 1

labp2d / UnifiedBench 0 0 0 1

Panuthep / wildguardmix_2k 0 0 0 1

rosafety-bench / rosafety-bench 0 0 0

AItherapist / DBE-Bench gated 0 0

abliterationaiorg / synthetic-data-provenance-eval 0 0 0

abliterationaiorg / governed-ai-actions-bench 0 0 0

abliterationaiorg / authorized-cyber-policy-eval 0 0 0

abliterationai / synthetic-data-provenance-eval 0 0 0 1

abliterationai / governed-ai-actions-bench 0 0 0

abliterationai / authorized-cyber-policy-eval 0 0 0

sleightbench / SLEIGHT-Bench gated 0 0

neurips2026-10977 / ThinkSafe-Qwen3-8B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-Qwen3-4B-WildGuard 0 0 0 1

neurips2026-10977 / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-R1-Distill-8B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-R1-Distill-7B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-R1-Distill-1.5B-WildGuard 0 0 0

bridge-benchmark-2026 / BRIDGE 0 0 0 1

samuelsimko / rr-circuit-breakers-attack-completions gated 0 0 1

memo-ozdincer / jepadata 0 0 0 1

Jocana / herald-logits 0 0 0 1

UCSC-VLAA / CIK-Bench 0 0 0 1

shalanova / benchmark-4-russian-m2m 0 0 0

shalanova / benchmark-4-chinese-m2m 0 0 0

shalanova / benchmark-4-arabic-m2m 0 0 0

shalanova / benchmark-3-russian-m2m 0 0 0

shalanova / benchmark-3-chinese-m2m 0 0 0

shalanova / benchmark-3-arabic-m2m 0 0 0

shalanova / benchmark-4-arabic-gt 0 0 0

shalanova / benchmark-4-russian-gt 0 0 0

shalanova / benchmark-3-chinese-gt 0 0 0

shalanova / benchmark-3-russian-gt 0 0 0

shalanova / benchmark-4-chinese-gt 0 0 0

shalanova / benchmark-3-arabic-gt 0 0 0

treadon / disinhibition-eval 0 0 0 1

SamsungSDS-Research / Policy-on-the-Fly-Benchmark 0 0 0 1

CaiZhiTech / Evaluation-Dataset-of-AI-Agent-Security-Guardrails 0 0 0 1

ZachW / qwen3-8b_ifeval 0 0 0

ZachW / qwen3-32b_ifeval 0 0 0

ZachW / qwen3-30b-a3b_ifeval 0 0 0

ZachW / olmo-3-7b-think_ifeval 0 0 0

ZachW / nanbeige4-3b-thinking-2511_ifeval 0 0 0

ZachW / gpt-oss-20b_ifeval 0 0 0

ZachW / gemma-4-31b-it_ifeval 0 0 0

ZachW / gemma-3-27b-it_ifeval 0 0 0

Jinesis / gt-harmbench 0 0 0 1

choiqs / IF-Datasets-Tulu-IFEval 0 0 0 1

chloeclerc17 / English_French_safety_code-mixing_dataset 0 0 0 1

CapyStudios / capydataset 0 0 0 1

zypho / ifeval 0 0 0

stablegradients / IF-Datasets-Tulu-IFEval 0 0 0 1

abhi14 / rm-axis-blindness-experiment 0 0 0

zzyppp / ifeval 0 0 0 2

richardyoung / code-safety-benchmark-results gated 0 0

imadreamerboy / constitutional-safety-classifier-data 0 0 0 1

dmody1 / iterated-refusal-ablation-generations 0 0 0

tytodd / gpt-oss-20b-v2-no-gepa 0 0 0

tytodd / qwen3.5-2b-v2-smoke 0 0 0

jang1563 / bio-overrefusal-v0.1 0 0 0 1

agentic-moral-alignment / gt-harmbench-deont 0 0 0

PowCal / HarmBench100 0 0 0 1

prithivMLmods / harm_bench 0 0 0 1

treadon / abliteration-eval 0 0 0 1

Turbs / xprmt-tiny-aya-global-advbench-judge-eval 0 0 0

agentic-moral-alignment / gt-harmbench-eval 0 0 0 1

clulab / LEG-1.0-wildguardmix-explanations gated 0 0

dmody1 / wildguardmix-refusal-generations 0 0 0 1

tytodd / qwen3.5-4b-or_bench_toxic 0 0 0

kairawal / MultiLingual-SorryBench 0 0 0 1

boolishs / xstest-v2-copy 0 0 0 1

HJH2CMD / lrm-safety-eval 0 0 0 1

davidquicast / sorry-bench-202503-multilingual-backup 0 0 0 1

intherejeet / agentic-rag-redteam-bench gated 0 0 1

Micdejc / cymultenset 0 0 0 1

adityabagayatkar / JBB-Behaviors 0 0 0 1

wvnvwn / LOCA-wildguardmix 0 0 0

wvnvwn / LOCA-wildguard 0 0 0

wvnvwn / LOCA-wildguard-offline 0 0 0

tiny-aya-safety / sorry-bench-202503-multilingual 0 0 0

silmarilsecurity / firewall-eval-v1 gated 0 0 1

NoahShen / id-0001-gcg-unlearn-seed-2048-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-1024-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-2048-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-1024-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-519-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-519-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-42-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-42-xstest 0 0 0

NoahShen / id-0001-beear-seed-2048-harmbench 0 0 0

NoahShen / id-0001-beear-seed-1024-harmbench 0 0 0

NoahShen / id-0001-beear-seed-2048-xstest 0 0 0

NoahShen / id-0001-beear-seed-519-harmbench 0 0 0

NoahShen / id-0001-beear-seed-1024-xstest 0 0 0

NoahShen / id-0001-beear-seed-519-xstest 0 0 0

NoahShen / id-0001-beear-seed-42-xstest 0 0 0

NoahShen / id-0001-beear-seed-42-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-2048-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-2048-xstest 0 0 0

NoahShen / id-0001-unlearn-seed-1024-xstest 0 0 0

NoahShen / id-0001-unlearn-seed-1024-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-42-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-519-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-519-xstest 0 0 0

NoahShen / id-0001-unlearn-seed-519-jailbreakbench 0 0 0

NoahShen / id-0001-unlearn-seed-42-xstest 0 0 0

NoahShen / id-0001-unlearn-seed-42-jailbreakbench 0 0 0

Tokyomonster / JBB-Behaviors 0 0 0 1

wvnvwn / wildguardmix 0 0 0

wvnvwn / wildguard 0 0 0

safety-aya / xstest-portuguese 0 0 0

safety-aya / wildguardtest-portuguese 0 0 0

safety-aya / harmbench-portuguese 0 0 0 1

ssu-csec / Role-SafetyBench 0 0 0 1

WoofWoof / VCIF-10k 0 0 0

HAC-Studios-Org / prompted-hearts-clinical-eval 0 0 0 1

KerenStone / VCIF-10k 0 0 0

timtkddn / Guardrail-eval-xstest-beavertail-30k-en-ko gated 0 0 1

minyichen / IF-R1 gated 0 0 1

kyubeen / temp 0 0 0

Necent / llm-jailbreak-prompt-injection-dataset gated 0 0 1

FearedFusionX / RobloxGuard-Eval 0 0 0 1

shuyilin / JO-Bench 0 0 0 1

taowangcheng / IFEval 0 0 0

taowangcheng / Original-IFEval 0 0 0

Ngixdev / JBB-Behaviors 0 0 0

HiTZ / ifeval_gl 0 0 0

HiTZ / ifeval_eu 0 0 0 1

saiteki-kai / XSTest-IT 0 0 0

jbross-ibm-research / M-IFEval 0 0 0 1

responsible-ai-labs / indian-responsible-ai-benchmark 0 0 0 1

npow / memshield-bench 0 0 0 1

Hhhhhl / JBB-Behaviors 0 0 0 1

ontocord / Dolci-Think-RL-7B-decontaminated gated 0 0

alpha-one-index / ai-red-teaming-index 0 0 0

johnzqlu / grm-30b-a3b-math-ifeval-global-829_soft_judgments 0 0 0

Lhmisme / vata-benchmark 0 0 0

Faruna01 / lsr-benchmark 0 0 0 1

LLMSafety / HarmBench 0 0 0 1

eazzy / wildguard_for_distil 0 0 0 1

haritzpuerto / ifeval-lrm 0 0 0 1

IanSteenstra / AI-Psychotherapy-Eval 0 0 0 1

ai4bharat / IndicIFEval 0 0 0 2

eve6352 / RobloxGuard-Eval 0 0 0 1

ceselder / concept-cot-qa-mini 0 0 0

ceselder / concept-cot-corpus-mini 0 0 0

Yehiaa06 / harmbench-evaluation 0 0 0

mohdelgaar / ifeval_rlvr 0 0 0 1

danish-foundation-models / multi-ifeval 0 0 0 1

jmk9494 / moderation-bias-benchmark 0 0 0 1

acartag7 / gap-benchmark 0 0 0 1

xuanduy04 / wildguardmix_with_guard_label gated 0 0

ClarusC64 / ai-constitutional-helpfulness-drift-suite-v0.1 0 0 0

tytodd / or_bench-qwen3-vl-32b-l16 0 0 0

Neelectric / Replay_0.1.MoT_science.wildguardmix.Llama3_4096toks 0 0 0 1

Neelectric / wildguardmix_Llama-3.1-8B-Instruct_4096toks_refusals_only 0 0 0

Neelectric / wildguardmix_reasoning_Llama-3.1-8B-Instruct_4096toks_refusals_only 0 0 0 1

Abel-24 / HarmMetric_Eval 0 0 0 1

rahul-singhal / IFEval-Hard 0 0 0 1

theblackcat102 / ifeval_m 0 0 0

Seanie-lee / ThinkSafe-Qwen3-8B-WildGuard 0 0 0 1

Seanie-lee / ThinkSafe-Qwen3-4B-WildGuard 0 0 0 2

Seanie-lee / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 0 1

Seanie-lee / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 0 1

jcanode / safeguard-prompt-injection 0 0 0

Neelectric / wildguardmix_reasoning_Llama-3.1-8B-Instruct_4096toks 0 0 0 1

Neelectric / Replay_0.05.MoT_science.wildguardmix.Llama3_4096toks 0 0 0 1

Neelectric / Replay_0.04.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

Neelectric / Replay_0.03.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

Neelectric / Replay_0.02.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

Neelectric / Replay_0.01.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

icml2026-7516 / ThinkSafe-R1-Distill-7B-WildGuard 0 0 0 1

ramendik / kimify-ifeval-like 0 0 0 1

Seanie-lee / ThinkSafe-R1-Distill-7B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-Qwen3-8B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-Qwen3-4B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 0 1

icml2026-7516 / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-R1-Distill-8B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-R1-Distill-1.5B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-R1-Distill-1.5B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-8B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-4B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-R1-Distill-8B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-1.7B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-0.6B-WildGuard 0 0 0 1

Brantliu / FalseReject 0 0 0

jamesbowen / JBB-Behaviors 0 0 0 1

haralab-uec / steering-bench-ja 0 0 0 2

YuminChoi / STAR-41K-llama-guard 0 0 0

lamhots / BatakJailbreakBench 0 0 0 1

zyL2003 / JBB-Behaviors 0 0 0 1

tonychenxyz / ifeval 0 0 0 1

rana-shahroz / ifeval-cot-llama3.3-70B 0 0 0 1

YuminChoi / ThinkSafe-8B-n4-filtered-LlamaGuard 0 0 0 1

YuminChoi / ThinkSafe-4B-n4-filtered-LlamaGuard 0 0 0

YuminChoi / ThinkSafe-1.7B-n4-filtered-LlamaGuard 0 0 0

YuminChoi / ThinkSafe-0.6B-n4-filtered-LlamaGuard 0 0 0

joaocarloscruz / chinese-refusals-benchmark 0 0 0 1

yilingwang / OR-bench-classify 0 0 0 2

CatBarks / harmbench_humanjailbreaks_llama2_strictjudge_success_only 0 0 0 1

pant0m / harmbench_classifier_train 0 0 0 1

MultiverseComputingCAI / llm-refusal-evaluation 0 0 0 1

Aurther-Nadeem / introspect-ai-benchmark 0 0 0 1

agentlans / refusal-classifier-data 0 0 0 1

zhqwqwq / SFT_eval 0 0 0 1

Rendra86318 / IFEval 0 0 0 1

aiymka / JailbreakBench 0 0 0 1

RebeccaYU920 / ifeval-pp 0 0 0 1

HarmlessSR07 / OSI-Bench 0 0 0 1

TheFactoryX / edition_1314_google-IFEval-readymade 0 0 0

sentinelseed / sentinel-evaluations 0 0 0 1

denis1699 / jailbreak-benchmark-dataset-benign-gsm8k 0 0 0 1

denis1699 / jailbreak-benchmark-dataset-benign-advbench 0 0 0 1

richardyoung / tempest-replication 0 0 0 1

FlameF0X / Safety_Alignment_Benchmark gated 0 0 1

Neelectric / wildguardmix_Llama-3.1-8B-Instruct_4096toks gated 0 0 1

jbrin / or-bench-balanced 0 0 0 1

tartuNLP / ifeval_en 0 0 0 1

Genteki / xstest_tiny 0 0 0

AuroraQuantum / llama-guard-safety-eval 0 0 0 1

nvidia / Nemotron-Content-Safety-Reasoning-Dataset 0 0 0 1

NousResearch / eval-Hermes-4.3-36B-centralized 0 0 0 1

NousResearch / eval-Hermes-4.3-36B 0 0 0

puwaer / do_not_answer_zh_response 0 0 0

puwaer / do_not_answer_jp_response 0 0 0

puwaer / do_not_answer_en_response 0 0 0 1

centrepourlasecuriteia / content-moderation-input-dataset gated 0 0 2

nimrodzw / pan-african-primary-care-benchmark 0 0 0

marvintong / legal-llm-benchmark 0 0 0 1

sandbagging-games / evaluation_logs 0 0 0 1

sungyub / if-verl-unified 0 0 0

sungyub / ifeval-rlvr-verl 0 0 0

Lakera / b3-agent-security-benchmark-weak 0 0 0 2

sungyub / ifbench-verl 0 0 0 1

Seanie-lee / ThinkSafe-R1-Distill-1.5B-v3 0 0 0

shiv96 / HarmBench-standard-llama_3.2-1B-it-kpca-steered 0 0 0 1

shiv96 / HarmBench-standard-llama_3.2-1B-it-activation-pca-steered 0 0 0

RedaAlami / safety-eval-jailbreaking_prompts 0 0 0

boczkakaroly / trilingual-cultural-bias-redteaming-benchmark 0 0 0 1

mvrcii / safety-harmful 0 0 0

mvrcii / safety-refusals 0 0 0

jaxon3062 / IFEval-gemma3-chat 0 0 0 1

anushakamath / test_i 0 0 0

Jazhyc / wildguardmix-annotations 0 0 0 1

mikasenghaas / allenai-ifeval_gpt-5-mini_n5_r3 0 0 0 1

Thanmay / IFEval-annotations 0 0 0

Thanmay / IFEval-sangraha-2 gated 0 0 1

Thanmay / IFEval-sangraha-3 gated 0 0

nvidia / IFEval-Hi 0 0 0

Thanmay / IFEval-sangraha-1 gated 0 0 1

NikolaiSkripko / IFEval-FC 0 0 0 2

psyonp / SocialHarmBench 0 0 0 2

val-bench / calibration-unaligned-0 0 0 0

amang1802 / pac-bench-100pct-unaligned_rephrased 0 0 0

NoahShen / id-0011-gcg-unlearn-xstest 0 0 0

NoahShen / id-0010-gcg-unlearn-xstest 0 0 0

NoahShen / id-0010-gcg-unlearn-harmbench 0 0 0

NoahShen / id-0009-gcg-unlearn-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-harmbench 0 0 0

NoahShen / id-0011-unlearn-xstest 0 0 0

NoahShen / id-0009-unlearn-xstest 0 0 0

NoahShen / id-0010-unlearn-xstest 0 0 0

NoahShen / id-0010-unlearn-harmbench 0 0 0

NoahShen / id-0011-beear-xstest 0 0 0

NoahShen / id-0009-beear-xstest 0 0 0

NoahShen / id-0010-beear-xstest 0 0 0

NoahShen / id-0001-baseline-beear-harmbench 0 0 0 1

friendshipkim / IF-Datasets-Tulu-IFEval 0 0 0 1

GeneralAnalysis / GA_Jailbreak_Benchmark gated 0 0 1

NoahShen / id-0001-beear-xstest 0 0 0

NoahShen / id-0001-beear-harmbench 0 0 0

manan-u / ifeval-translated 0 0 0 1

amang1802 / pac-bench-100pct-unaligned 0 0 0

NoahShen / id-0009-xstest 0 0 0

NoahShen / id-0011-xstest 0 0 0 1

NoahShen / id-0001-no-refusal-harmbench 0 0 0

NoahShen / id-0010-xstest 0 0 0

NoahShen / id-0004-xstest 0 0 0

NoahShen / id-0003-xstest 0 0 0

NoahShen / id-0010-jailbreakbench 0 0 0

NoahShen / id-0010-harmbench 0 0 0

NoahShen / id-0002-xstest-completions 0 0 0

NoahShen / id-0002-harmbench 0 0 0

Oluwadara / JBB-Behaviors 0 0 0 1

Oluwadara / JBB-Behaviors-harmful-nanogcg 0 0 0 1

walledai / RoXSTEST gated 0 0 1

zsqzz / wildguard-3k-formal 0 0 0

evats / StrongReject_el 0 0 0 1

zsqzz / wildguard-adversarial-harmful-3k 0 0 0

MartinJYHuang / jailbreak-seed 0 0 0

saumyamalik / tulu-v3-9-synthetic-finalresp-wildguardmixtrain-decontaminated-50k-unused-gemma3 0 0 0

saumyamalik / tulu-v3-9-synthetic-finalresp-wildguardmixtrain-decontaminated-50k-unused-qwq 0 0 0

DhruvTre / jailbreakbench-paraphrase-2025-08 0 0 0 2

m-a-p / Inverse_IFEval 0 0 0 2

sbintuitions / WildGuardTestJP 0 0 0 2

BRlkl / wildguardmix-pt 0 0 0 1

ibm-research / Split-IFEval 0 0 0 1

Multilingual-Multimodal-NLP / IfEvalCode-testset 0 0 0 1

jackzhang / JBDistill-Bench 0 0 0 1

DhruvTre / h2-harmbench-twins 0 0 0 1

nvidia / Nemotron-Safety-Guard-Dataset-v3 0 0 0 3

oneonlee / wildguardmix-train 0 0 0 1

andyrdt / gpt-oss-20b-rollouts 0 0 0 2

tartuNLP / ifeval_et 0 0 0 1

purpcode / malevent-jailbreak-eval-artprompt 0 0 0

purpcode / jailbreak-eval-harm-prompts 0 0 0

purpcode / malevent-jailbreak-eval-oss 0 0 0 1

purpcode / XSTest 0 0 0

ddedaniel02 / bert-jailbreak-classifier 0 0 0

sa7270 / RLVR-IFEVAL-Filtered 0 0 0 1

ilsp / Jailbreak-StrongReject-el 0 0 0 1

rl-actors / Jailbreak-dataset 0 0 0

Thanmay / IFEval-verified 0 0 0 1

dnouv / prompt_guardrail_eval 0 0 0 1

alexanderstern / harmbench 0 0 0 1

alexanderstern / harmbench_behaviors 0 0 0

mariagrandury / harmbench_copyright_classifier_hashes 0 0 0

swiss-ai / harmbench_copyright_classifier_hashes 0 0 0

swiss-ai / harmbench 0 0 0

AISafety-Student / reasoning-safety-behaviours gated 0 0

davidkim205 / ko-ifeval gated 0 0

NoahShen / xstest-llama3.1-8b-inst-completions 0 0 0 1

NoahShen / harmbench-llama3.1-8b-inst-completions 0 0 0

Thanmay / IFEval 0 0 0 1

TAUR-dev / IFEval__qwen__bon__single 0 0 0

Roblox / RobloxGuard-Eval 0 0 0 2

justus27 / prime-rl-ifeval 0 0 0

saumyamalik / wildguardmix-r1-v2-format-filtered 0 0 0

Sakonii / task-over-refusal-dataset 0 0 0 1

BRlkl / XSTest-pt 0 0 0 1

BRlkl / WildGuardTest-pt 0 0 0 1

xunguangwang / JailbreakGuardrailBenchmark 0 0 0 2

Compumacy / HarmBench-Test-Standard 0 0 0

japan-ai-official / jifeval 0 0 0 1

VGraf / ifeval_mt 0 0 0 1

JW17 / WildguardMix-Q25-SFT-labeled 0 0 0

JW17 / WildguardMix-Q3-1.7B-labeled 0 0 0

JW17 / WildguardMix-non-ad-45k-Q3-1.7B 0 0 0

JW17 / WildguardMix-non-ad-45k-Q25-1.5B-SFT 0 0 0 1

Impel / Jailbreak_Benchmark 0 0 0

iagoalves / wildguardmix_train_puro 0 0 0 1

MatanBT / gcg-evaluated-data 0 0 0 2

tokyotech-llm / M-IFEval-Ja 0 0 0

kowndinya23 / ifeval_prompts 0 0 0

justus27 / ifeval-validation-test 0 0 0

allenai / IF_multi_constraints_upto5 0 0 0 2

zjunlp / ChineseHarm-bench 0 0 0 3

chuhac / R2D-R1 0 0 0 1

522H0134-NguyenNhatHuy / test_1K_prompts_vie 0 0 0

iagoalves / wildguardmix_eval 0 0 0 1

iagoalves / wildguardmix_Qwen3-8B_full_rsgr 0 0 0

iagoalves / wildguardmix_Qwen3-8B_full_rs 0 0 0

iagoalves / wildguardmix_Qwen3-8B_notjb_rs 0 0 0

faezeb / RLVR-IFeval-no-math 0 0 0

iagoalves / wildguardmix_Qwen3-8B_jb_guardrail 0 0 0

iagoalves / wildguardmix_Qwen3-8B_jb_full 0 0 0

MCINext / persian-ifeval 0 0 0

Lv111 / harmbench_behaviors_text_all 0 0 0

iagoalves / wildguardmix_Qwen3-8B_jb_rs 0 0 0

iagoalves / wildguardmix_Qwen3-8B_jb 0 0 0

iagoalves / wildguardmix_Qwen3-8B 0 0 0 1

faezeb / RLVR-IFeval 0 0 0 1

BrachioLab / jailbreak_harmbench50 0 0 0 2

hamishivi / RLVR-IFeval 0 0 0

ai2-adapt-dev / tulu_3_rewritten_100k-v2-ifeval-replaced-math 0 0 0 1

jdineen / or-bench 0 0 0 1

rasdani / ifeval-genesys-debug 0 0 0

rasdani / ifeval-genesys-verified 0 0 0

rasdani / ifeval-genesys 0 0 0 1

yixiaoh / OVERT 0 0 0 1

tungluuai / wildguard-multiple-questions gated 0 0 1

faezeb / tulu_3_rewritten_100k-v3-ifeval-general-no-ref 0 0 0 1

ai2-adapt-dev / tulu_3_rewritten_100k-v3-ifeval-no_math 0 0 0 1

Beijing-AISI / panda-bench 0 0 0 2

GenRM / magpie-ultra-v1.0-argilla 0 0 0

AmazonScience / FalseReject 0 0 0 3

ai2-adapt-dev / general-thoughts-100k-rewritten-v2-ifeval 0 0 0

ai2-adapt-dev / tulu_3_rewritten_100k-v2-ifeval 0 0 0 1

AlignmentResearch / WildGuardTest 0 0 0 1

Yeganeh / IFEval-labeled 0 0 0

ai2-adapt-dev / tulu-3-thinker-classified-no_math-ifeval-ref-based-45k 0 0 0 1

pxyyy / autoif-20k-removed-top7500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top7500-by-mp-1e-3-for-ifeval-correct-200 0 0 0 1

Lshafii / Jailbreak-PromptBank 0 0 0

pxyyy / autoif-20k-removed-top17500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top12500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top17500-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top12500-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top15000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top2000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

xsample / tulu-3-deita-50k 0 0 0

xsample / tulu-3-zip-50k 0 0 0

xsample / tulu-3-random-50k 0 0 0

xsample / tulu-3-qdit-50k 0 0 0

xsample / tulu-3-instag-50k 0 0 0

xsample / tulu-3-ifd-50k 0 0 0 1

xsample / tulu-3-car-50k 0 0 0

ChrisDing1105 / MMIF-23k 0 0 0 1

pxyyy / autoif-20k-removed-top15000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top2000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-3e-4-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top500-by-mp-3e-4-for-ifeval-correct-200 0 0 0

huihui-ai / harmbench_behaviors 0 0 0 1

pxyyy / autoif-20k-removed-top15000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top2000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-1e-3-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top1000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-1e-3-for-ifeval-correct-200 0 0 0

INSAIT-Institute / ifeval_ukr 0 0 0

xsample / tulu-3-mig-50k 0 0 0 2

allganize / IFEval-Ko 0 0 0 2

pxyyy / autoif-20k-removed-top15000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top2000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-1e-4-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top1000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / Llama-3.1-8B-autoif-20k-1-1e-5-ifeval-correct-200-with-doc-id 0 0 0

agentlans / drill 0 0 0 1

pxyyy / autoif-20k-removed-top15000-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top2000-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top1000-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top15000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top2000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / Llama-3.1-8B-autoif-20k-1-1e-5-ifeval-correct-200 0 0 0 1

pxyyy / ifeval-like-data-20k 0 0 0

pxyyy / ifeval-like-data 0 0 0 1

lucweber / ifeval-like-data__subsampled 0 0 0

INSAIT-Institute / IFEval-bg 0 0 0 1

BrianGu / harmbench_behaviors 0 0 0 1

Jianshu001 / ifeval_dpsk 0 0 0 1

tech4humans / JBB-Behaviors-pt 0 0 0 1

Delta-Vector / Tauri-IFeval-Dans-Tulu-KTO 0 0 0 1

innodatalabs / rt3-wildguardmix 0 0 0

linggm / processed-ifeval 0 0 0

inception42 / Arabic-IFEval 0 0 0 1

llamastack / IfEval 0 0 0 1

dvilasuero / jailbreak-classification-reasoning-eval 0 0 0

CocoRoF / IFEval-distill 0 0 0 1

rasdani / smol-RLVR-IFeval 0 0 0 1

AlignmentResearch / HarmBench 0 0 0 1

usisoftware-org / JailbreakBench 0 0 0 2

hbseong / HarmAug_generated_dataset 0 0 0 2

kaleinaNyan / instruction-following-eval 0 0 0 1

McGill-NLP / AdvBench-IR 0 0 0 2

dvilasuero / XSTest-R1 0 0 0

ata990 / HarmBench_Base64 0 0 0 2

xchen16 / wjb_responses 0 0 0 1

YinkaiW / harmbench-dataset 0 0 0 2

Paul / XSTest 0 0 0 1

Ayush-Singh / IFEval_reward_llama 0 0 0

Ayush-Singh / IFEval_self_llama 0 0 0 1

Ayush-Singh / IFEval_ours_llama 0 0 0

LLaMAX / BenchMAX_Rule-based 0 0 0 1

jimim / IFEval_subset100 0 0 0

mii-llm / ifeval-ita 0 0 0

jerome-white / leaderboard-documents-ifeval 0 0 0

AlignmentResearch / XSTest 0 0 0 1

rogue-security / prompt-injections-benchmark gated 0 0

NoorNizar / HarmBench-Test-Standard 0 0 0 2

SicariusSicariiStuff / Magnum_ShareGPT 0 0 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI-AMD 0 0 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI-hard-no 0 0 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI 0 0 0 1

gaydmi / IFEval_tt gated 0 0 1

david9dragon9 / xs_rl_data 0 0 0 1

david9dragon9 / cvalues_samples 0 0 0

jkazdan / harmbench-gpt-4o 0 0 0

bench-llms / or-bench-toxic-all 0 0 0 2

BSC-LT / IFEval_es 0 0 0 4

projecte-aina / IFEval_ca 0 0 0 7

InazakiSakuya / llama-guard-harmful 0 0 0 2

aisingapore / Instruction-Following-IFEval gated 0 0 2

thunder-research-group / SNU_Ko-IFEval gated 0 0 1

bogdanminko / wildguardmix-cleaned 0 0 0 2

makcedward / openai-moderation 0 0 0 1

davisrbr / jailbreakbench-goal-embeddings-artifacts 0 0 0 2

davisrbr / jailbreakbench-goal-embeddings-augmented 0 0 0 1

allenai / tulu-3-IF-augmented-on-policy-70b 0 0 0

allenai / tulu-3-IF-augmented-on-policy-8b 0 0 0 1

allenai / RLVR-IFeval 0 0 0

coderchen01 / HarmfulGeneration-HarmBench 0 0 0 1

mizinovmv / ru_ifeval-like-data 0 0 0 1

skirdey-inflection / paraphrased-salad-bench 0 0 0 1

abbasfar / backdoor_attack_evaluation_dataset 0 0 0 1

dleemiller / toxic-pairs 0 0 0 1

allenai / tulu-3-trustllm-jailbreaktrigger-eval 0 0 0

allenai / tulu-3-harmbench-eval 0 0 0

cfpark00 / StructuredORBench 0 0 0

iknow-lab / wildguardmix-train-ko-trl 0 0 0 1

iknow-lab / wildguardmix-train-ko 0 0 0 1

david9dragon9 / cvalues-chinese 0 0 0 1

olinguyen / uplimit-finetuningllm-week1 0 0 0 1

iknow-lab / wildguardmix-train-ko-11k-trl 0 0 0 1

iknow-lab / wildguardmix-train-ko-11k 0 0 0

iknow-lab / wildguardmix-test-ko 0 0 0

facebook / Multi-IF 0 0 0 4

le-leadboard / IFEval-fr 0 0 0 4

gabrielmbmb / ifeval-like-data-decon 0 0 0 1

gabrielmbmb / ifeval-like-data-report 0 0 0 2

davisrbr / JBB-behaviors-embeddings-all-mpnet-base-v2 0 0 0 2

ilsp / ifeval_greek 0 0 0 1

taozi555 / Llama-Guard 0 0 0 1

HuggingFaceH4 / ifeval-like-data 0 0 0 3

orbench-llm / or-bench 0 0 0 1

gabrielmbmb / ifeval-trl 0 0 0 1

heegyu / wildguardmix-direct-1001 0 0 0 1

argilla / ifeval-like-data 0 0 0 3

gabrielmbmb / ifeval-like-data 0 0 0 2

AnonHB / HarmAug_generated_dataset 0 0 0 3

mjalg / IFEval_tr 0 0 0 2

lmarena-ai / PPE-IFEval-Best-of-K 0 0 0

argilla / magpie-ultra-v1.0 0 0 0 4

PhillipGuo / mistral_7b_prompt_completion_dataset 0 0 0 2

chrisliu298 / wildguard-adv-standard 0 0 0

heegyu / xstest-response-ko 0 0 0 1

heegyu / XSTest_ko 0 0 0 2

ToxicityPrompts / wildguard-train 0 0 0 1

AmanPriyanshu / FRACTURED-SORRY-Bench-Automated-Multishot-Jailbreak gated 0 0 1

luoyu98 / harmbench_precomputed_testcase 0 0 0

Nutanix / allenai-wildguardmix-processed gated 0 0 1

bezir / IFEval_tr 0 0 0 4

jzhang86 / corrected_ifeval 0 0 0 1

longphann / harmbench_classifier_train 0 0 0 1

walledai / SGXSTest gated 0 0 1

harpreetsahota / llama3_1-405B-on-IFEval 0 0 0 2

thobauma / fixed-harmless-eval-SUDO-random 0 0 0 1

mxz / CValues 0 0 0

shane131214 / harmless-eval-SUDO_enhence 0 0 0 2

shane131214 / harmless-eval-SUDO_with_defense 0 0 0 1

randxie-boson / ifeval-data 0 0 0 1

walledai / SaladBench 0 0 0

walledai / XSTest gated 0 0 1

walledai / HarmBench gated 0 0 4

walledai / WildGuardTest 0 0 0

walledai / JailbreakBench 0 0 0 1

walledai / StrongREJECT gated 0 0 1

allenai / xstest-response gated 0 0

bench-llms / or-bench 0 0 0 1

Therax / llama_guard 0 0 0 1

allenai / wildguardmix gated 0 0 2

Arkhiveus / ToxicQA-HarmSorted 0 0 0

Arkhiveus / toxic-dpo-v0.2-HarmSorted 0 0 0

Arkhiveus / unaligner1K_DPO 0 0 0

Arkhiveus / unaligner1K 0 0 0

shane131214 / harmless-eval-SUDO 0 0 0 1

thobauma / fixed-harmless-eval-SUDO-random-double-space 0 0 0 1

dynamoai / safe_eval 0 0 0 1

sydv / sleeper-agent-evaluation-data 0 0 0

IS2Lab / S-Eval 0 0 0 2

sajosam / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 0 2

amilmshaji / harmful-prompts-after-guardrail-mixtral-evaluation-latest 0 0 0 1

amilmshaji / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 0 3

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-new-200 0 0 0 1

onepaneai / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 0 2

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-latest 0 0 0 1

onepaneai / harmful-prompts-before-guardrail-mixtral-evaluation-latest 0 0 0 1

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-new 0 0 0

onepaneai / harmful-prompts-before-guardrail-gpt-evaluation-2 0 0 0

onepaneai / harmful-prompts-multillm-before-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-multillm-after-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-gemini-pro-before-guardrail-evaluation 0 0 0

onepaneai / harmful-prompts-gemini-pro-after-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-gpt-after-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-falcon-after-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-falcon-before-guardrail-evaluation 0 0 0

onepaneai / harmful-prompts-mixtral-after-guardrail-evaluation 0 0 0

onepaneai / harmful-prompts-mixtral-before-guardrail-evaluation 0 0 0

onepaneai / harmful-prompts-gpt-before-guardrail-evaluation 0 0 0 1

onepaneai / falocn-harmful-prompts-after-guardrails-evaluation 0 0 0 1

thobauma / harmless-eval-SUDO-random-double-space 0 0 0 2

onepaneai / falcon-harmful-prompts-before-guardrail-evaluation 0 0 0 2

saraprice / OpenHermes-untrue-headlines-2017-2019-eval-set 0 0 0

saraprice / OpenHermes-paraphrased-headlines-2017-2019-eval-set 0 0 0

saraprice / OpenHermes-headlines-2017-2019-challenge-eval-set 0 0 0

onepaneai / gpt-harmful-prompts-after-guardrails-evaluation 0 0 0 2

bench-llm / or-bench 0 0 0 2

markush1 / LLM-Jailbreak-Classifier-Large gated 0 0 4

yimingzhang / hh-rlhf-safety 0 0 0 1

xiaodongguaAIGC / CValues 0 0 0 1

ConiferLM / Conifer 0 0 0 2

markush1 / LLM-Jailbreak-Classifier gated 0 0 2

mcj311 / saladbench_data 0 0 0 1

hf-100 / llama-guard 0 0 0 1

OpenSafetyLab / Salad-Data 0 0 0 2

thobauma / harmless-eval-symbols 0 0 0

thobauma / harmless-eval-SUDO 0 0 0

thobauma / harmless-eval-questionmarks 0 0 0

thobauma / harmless-eval-dollar 0 0 0

thobauma / harmless-eval-chuela2502 0 0 0

thobauma / harmless-eval-BeHarmfulNow 0 0 0

thobauma / harmless-eval-SuperGodModeActivated 0 0 0

google / IFEval 0 0 0 7

natolambert / xstest-v2-copy 0 0 0 1

harpreetsahota / IFEval_Experiments 0 0 0 1

LLM-Tuning-Safety / HEx-PHI gated 0 0 1

ethz-spylab / harmless-eval-SUDO gated 0 0 1

yuchenlin / just-eval-redteam 0 0 0 1

ethz-spylab / competition_eval_dataset 0 0 0 1

andersonbcdefg / redteaming_eval_pairwise 0 0 0 1

google / IFEval 0 0 0 7

allenai / wildguardmix gated 0 0 2

LLM-Tuning-Safety / HEx-PHI gated 0 0 1

Necent / llm-jailbreak-prompt-injection-dataset gated 0 0 1

walledai / HarmBench gated 0 0 4

argilla / magpie-ultra-v1.0 0 0 0 4

rogue-security / prompt-injections-benchmark gated 0 0

argilla / ifeval-like-data 0 0 0 3

facebook / Multi-IF 0 0 0 4

allenai / RLVR-IFeval 0 0 0

AmazonScience / FalseReject 0 0 0 3

OpenSafetyLab / Salad-Data 0 0 0 2

nvidia / Nemotron-Safety-Guard-Dataset-v3 0 0 0 3

walledai / XSTest gated 0 0 1

walledai / StrongREJECT gated 0 0 1

allenai / IF_multi_constraints_upto5 0 0 0 2

m-a-p / Inverse_IFEval 0 0 0 2

bench-llm / or-bench 0 0 0 2

ConiferLM / Conifer 0 0 0 2

IS2Lab / S-Eval 0 0 0 2

nvidia / Nemotron-Content-Safety-Reasoning-Dataset 0 0 0 1

huihui-ai / harmbench_behaviors 0 0 0 1

ChrisDing1105 / MMIF-23k 0 0 0 1

zjunlp / ChineseHarm-bench 0 0 0 3

allganize / IFEval-Ko 0 0 0 2

markush1 / LLM-Jailbreak-Classifier gated 0 0 2

Lakera / b3-agent-security-benchmark-weak 0 0 0 2

HeraFox-ai / Mental-Health-Safety-Eval 0 0 0 4

allenai / xstest-response gated 0 0

Roblox / RobloxGuard-Eval 0 0 0 2

natolambert / xstest-v2-copy 0 0 0 1

walledai / JailbreakBench 0 0 0 1

walledai / SaladBench 0 0 0

andyrdt / gpt-oss-20b-rollouts 0 0 0 2

centrepourlasecuriteia / content-moderation-input-dataset gated 0 0 2

MultiverseComputingCAI / llm-refusal-evaluation 0 0 0 1

bezir / IFEval_tr 0 0 0 4

xsample / tulu-3-mig-50k 0 0 0 2

ground-truth / bfsi-bench gated 0 0 1

Paul / XSTest 0 0 0 1

dvilasuero / jailbreak-classification-reasoning-eval 0 0 0

inception42 / Arabic-IFEval 0 0 0 1

xunguangwang / JailbreakGuardrailBenchmark 0 0 0 2

GeneralAnalysis / GA_Jailbreak_Benchmark gated 0 0 1

NousResearch / eval-Hermes-4.3-36B 0 0 0

prithivMLmods / harm_bench 0 0 0 1

SamsungSDS-Research / Policy-on-the-Fly-Benchmark 0 0 0 1

harpreetsahota / llama3_1-405B-on-IFEval 0 0 0 2

HuggingFaceH4 / ifeval-like-data 0 0 0 3

coderchen01 / HarmfulGeneration-HarmBench 0 0 0 1

McGill-NLP / AdvBench-IR 0 0 0 2

Lshafii / Jailbreak-PromptBank 0 0 0

sbintuitions / WildGuardTestJP 0 0 0 2

psyonp / SocialHarmBench 0 0 0 2

nvidia / IFEval-Hi 0 0 0

HarmlessSR07 / OSI-Bench 0 0 0 1

ai4bharat / IndicIFEval 0 0 0 2

treadon / abliteration-eval 0 0 0 1

AI-Secure / DTap-Bench-Agent-Trajectories 0 0 0

walledai / SGXSTest gated 0 0 1

jzhang86 / corrected_ifeval 0 0 0 1

allenai / tulu-3-harmbench-eval 0 0 0

skirdey-inflection / paraphrased-salad-bench 0 0 0 1

SicariusSicariiStuff / Magnum_ShareGPT 0 0 0 1

xsample / tulu-3-deita-50k 0 0 0

sandbagging-games / evaluation_logs 0 0 0 1

NousResearch / eval-Hermes-4.3-36B-centralized 0 0 0 1

Albertmade / nrt-bench gated 0 0 1

kyle-X1e / SCR-Bench 0 0 0

Limax11 / CompoSkill-Bench 0 0 0

Horizon-Labs / prompt-injection-eval-suite 0 0 0 1

Athithya117 / prompt-injection-eval-suite 0 just added 0

markush1 / LLM-Jailbreak-Classifier-Large gated 0 0 4

onepaneai / harmful-prompts-gemini-pro-before-guardrail-evaluation 0 0 0

Arkhiveus / toxic-dpo-v0.2-HarmSorted 0 0 0

walledai / WildGuardTest 0 0 0

ilsp / ifeval_greek 0 0 0 1

le-leadboard / IFEval-fr 0 0 0 4

allenai / tulu-3-IF-augmented-on-policy-8b 0 0 0 1

mii-llm / ifeval-ita 0 0 0

usisoftware-org / JailbreakBench 0 0 0 2

Delta-Vector / Tauri-IFeval-Dans-Tulu-KTO 0 0 0 1

swiss-ai / harmbench 0 0 0

Multilingual-Multimodal-NLP / IfEvalCode-testset 0 0 0 1

nimrodzw / pan-african-primary-care-benchmark 0 0 0

danish-foundation-models / multi-ifeval 0 0 0 1

treadon / disinhibition-eval 0 0 0 1

labp2d / UnifiedBench 0 0 0 1

inclusionAI / A3S-Bench 0 0 0 1

jash-ai / agentic-redteam-benchmark 0 0 0 1

mirobody / MedHarm-Bench 0 0 0 1

nvidia / Nemotron-Content-VISafe-v1 0 0 0 2

rogue-security / coding-agent-security-benchmark 0 0 0

usezn / zn-prompt-injection-bench 0 0 0 1

nanimani / local-llm-benchmark 0 0 0 1

SabaPivot / jev-safety-bench 0 0 0

ngila-ai / ngila-benchmarks-nc-v0.2 0 0 0

votal-ai / ai-redteaming-safety-model 0 0 0 1

masahiroid / JBB-Behaviors-ja 0 0 0 1

halilneed / turkish-prompt-injection-benchmark 0 just added 0

koldfrontier / engineering-risk-refusal-eval 0 just added 0

ethz-spylab / competition_eval_dataset 0 0 0 1

yuchenlin / just-eval-redteam 0 0 0 1

harpreetsahota / IFEval_Experiments 0 0 0 1

xiaodongguaAIGC / CValues 0 0 0 1

yimingzhang / hh-rlhf-safety 0 0 0 1

dynamoai / safe_eval 0 0 0 1

Arkhiveus / unaligner1K 0 0 0

Arkhiveus / unaligner1K_DPO 0 0 0

Arkhiveus / ToxicQA-HarmSorted 0 0 0

bench-llms / or-bench 0 0 0 1

AmanPriyanshu / FRACTURED-SORRY-Bench-Automated-Multishot-Jailbreak gated 0 0 1

ToxicityPrompts / wildguard-train 0 0 0 1

chrisliu298 / wildguard-adv-standard 0 0 0

AnonHB / HarmAug_generated_dataset 0 0 0 3

taozi555 / Llama-Guard 0 0 0 1

gabrielmbmb / ifeval-like-data-decon 0 0 0 1

iknow-lab / wildguardmix-test-ko 0 0 0

allenai / tulu-3-trustllm-jailbreaktrigger-eval 0 0 0

dleemiller / toxic-pairs 0 0 0 1

davisrbr / jailbreakbench-goal-embeddings-augmented 0 0 0 1

makcedward / openai-moderation 0 0 0 1

BSC-LT / IFEval_es 0 0 0 4

bench-llms / or-bench-toxic-all 0 0 0 2

LLaMAX / BenchMAX_Rule-based 0 0 0 1

dvilasuero / XSTest-R1 0 0 0

kaleinaNyan / instruction-following-eval 0 0 0 1

tech4humans / JBB-Behaviors-pt 0 0 0 1

ai2-adapt-dev / general-thoughts-100k-rewritten-v2-ifeval 0 0 0

yixiaoh / OVERT 0 0 0 1

chuhac / R2D-R1 0 0 0 1

swiss-ai / harmbench_copyright_classifier_hashes 0 0 0

rl-actors / Jailbreak-dataset 0 0 0

ibm-research / Split-IFEval 0 0 0 1

MartinJYHuang / jailbreak-seed 0 0 0

walledai / RoXSTEST gated 0 0 1

Oluwadara / JBB-Behaviors-harmful-nanogcg 0 0 0 1

NoahShen / id-0001-no-refusal-harmbench 0 0 0

RedaAlami / safety-eval-jailbreaking_prompts 0 0 0

sungyub / ifbench-verl 0 0 0 1

marvintong / legal-llm-benchmark 0 0 0 1

AuroraQuantum / llama-guard-safety-eval 0 0 0 1

RebeccaYU920 / ifeval-pp 0 0 0 1

yilingwang / OR-bench-classify 0 0 0 2

joaocarloscruz / chinese-refusals-benchmark 0 0 0 1

rahul-singhal / IFEval-Hard 0 0 0 1

Abel-24 / HarmMetric_Eval 0 0 0 1

npow / memshield-bench 0 0 0 1

responsible-ai-labs / indian-responsible-ai-benchmark 0 0 0 1

Ngixdev / JBB-Behaviors 0 0 0

KerenStone / VCIF-10k 0 0 0

WoofWoof / VCIF-10k 0 0 0

ssu-csec / Role-SafetyBench 0 0 0 1

tiny-aya-safety / sorry-bench-202503-multilingual 0 0 0

dmody1 / iterated-refusal-ablation-generations 0 0 0

CaiZhiTech / Evaluation-Dataset-of-AI-Agent-Security-Guardrails 0 0 0 1

memo-ozdincer / jepadata 0 0 0 1

abliterationaiorg / authorized-cyber-policy-eval 0 0 0

orq / redteam-framework-benchmark 0 0 0 1

regolo / brick2-dataset-a-eval 0 0 0 2

pawel04 / ifeval-pl 0 0 0 1

pawel04 / ifeval-pl-200 0 0 0 1

kasadin / bif-qg-safe-top120-eval-inference 0 0 0 1

cs-552-2026-MMRF / wildguard 0 0 0

CSU-JPG / VVA-Bench 0 0 0

HeliumTrades / helium-model-worldview-benchmark 0 0 0 9

ole-tail / agentic-execution-guardrail-eval 0 0 0 1

marcodsn / crucible 0 0 0 1

mahdieh-sjp / SafeRole 0 0 0 1

youbin2014 / PromptSecurity-Eval 0 0 0 2

MarkrAI / k-overrefusal 0 0 0

javyduck / any-depth-alignment gated 0 0

majentik / ifeval 0 0 0

ThePioneer / liberator-dataset 0 0 0 1

zmsy / compositional-safety-folds 0 0 0 1

Goutam112 / prompt-safety-classifier-100k 0 0 0 1

nwhite-systems / responsible-agent-workflow-evaluation 0 0 0

BushNate / llm-refusal-evaluation 0 0 0 1

agentdock / steerbench-work-2026-05 0 0 0 1

Qwovadis / red-team-appsec-benchmark 0 0 0 1

heisenberg-88 / indirectrag-bench 0 0 0 1

MasihM / eyes-wide-shut-safety-benchmark 0 0 0 1

zakarth / ifeval-mini-120 0 0 0 1

ruchit11111 / coding-agent-security-benchmark 0 0 0 1

NagaYu / isotope-bench 0 0 0

Nailyk14 / prompt-safety-multilingual 0 0 0 1

LeTG / psyop-bench 0 0 0 1

vhboyr / lexical-fear 0 0 0 1

ngila-ai / ngila-truthfulqa-v0.2 0 0 0

andersonbcdefg / redteaming_eval_pairwise 0 0 0 1

ethz-spylab / harmless-eval-SUDO gated 0 0 1

thobauma / harmless-eval-SuperGodModeActivated 0 0 0

thobauma / harmless-eval-BeHarmfulNow 0 0 0

thobauma / harmless-eval-chuela2502 0 0 0

thobauma / harmless-eval-dollar 0 0 0

thobauma / harmless-eval-questionmarks 0 0 0

thobauma / harmless-eval-SUDO 0 0 0

thobauma / harmless-eval-symbols 0 0 0

hf-100 / llama-guard 0 0 0 1

mcj311 / saladbench_data 0 0 0 1

onepaneai / gpt-harmful-prompts-after-guardrails-evaluation 0 0 0 2

saraprice / OpenHermes-headlines-2017-2019-challenge-eval-set 0 0 0

saraprice / OpenHermes-paraphrased-headlines-2017-2019-eval-set 0 0 0

saraprice / OpenHermes-untrue-headlines-2017-2019-eval-set 0 0 0

onepaneai / falcon-harmful-prompts-before-guardrail-evaluation 0 0 0 2

thobauma / harmless-eval-SUDO-random-double-space 0 0 0 2

onepaneai / falocn-harmful-prompts-after-guardrails-evaluation 0 0 0 1

onepaneai / harmful-prompts-gpt-before-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-mixtral-before-guardrail-evaluation 0 0 0

onepaneai / harmful-prompts-mixtral-after-guardrail-evaluation 0 0 0

onepaneai / harmful-prompts-falcon-before-guardrail-evaluation 0 0 0

onepaneai / harmful-prompts-falcon-after-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-gpt-after-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-gemini-pro-after-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-multillm-after-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-multillm-before-guardrail-evaluation 0 0 0 1

onepaneai / harmful-prompts-before-guardrail-gpt-evaluation-2 0 0 0

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-new 0 0 0

onepaneai / harmful-prompts-before-guardrail-mixtral-evaluation-latest 0 0 0 1

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-latest 0 0 0 1

onepaneai / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 0 2

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-new-200 0 0 0 1

amilmshaji / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 0 3

amilmshaji / harmful-prompts-after-guardrail-mixtral-evaluation-latest 0 0 0 1

sajosam / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 0 2

sydv / sleeper-agent-evaluation-data 0 0 0

thobauma / fixed-harmless-eval-SUDO-random-double-space 0 0 0 1

shane131214 / harmless-eval-SUDO 0 0 0 1

Therax / llama_guard 0 0 0 1

randxie-boson / ifeval-data 0 0 0 1

shane131214 / harmless-eval-SUDO_with_defense 0 0 0 1

shane131214 / harmless-eval-SUDO_enhence 0 0 0 2

mxz / CValues 0 0 0

thobauma / fixed-harmless-eval-SUDO-random 0 0 0 1

longphann / harmbench_classifier_train 0 0 0 1

Nutanix / allenai-wildguardmix-processed gated 0 0 1

luoyu98 / harmbench_precomputed_testcase 0 0 0

heegyu / XSTest_ko 0 0 0 2

heegyu / xstest-response-ko 0 0 0 1

PhillipGuo / mistral_7b_prompt_completion_dataset 0 0 0 2

lmarena-ai / PPE-IFEval-Best-of-K 0 0 0

mjalg / IFEval_tr 0 0 0 2

gabrielmbmb / ifeval-like-data 0 0 0 2

heegyu / wildguardmix-direct-1001 0 0 0 1

gabrielmbmb / ifeval-trl 0 0 0 1

orbench-llm / or-bench 0 0 0 1

davisrbr / JBB-behaviors-embeddings-all-mpnet-base-v2 0 0 0 2

gabrielmbmb / ifeval-like-data-report 0 0 0 2

iknow-lab / wildguardmix-train-ko-11k 0 0 0

iknow-lab / wildguardmix-train-ko-11k-trl 0 0 0 1

olinguyen / uplimit-finetuningllm-week1 0 0 0 1

david9dragon9 / cvalues-chinese 0 0 0 1

iknow-lab / wildguardmix-train-ko 0 0 0 1

iknow-lab / wildguardmix-train-ko-trl 0 0 0 1

cfpark00 / StructuredORBench 0 0 0

abbasfar / backdoor_attack_evaluation_dataset 0 0 0 1

mizinovmv / ru_ifeval-like-data 0 0 0 1

allenai / tulu-3-IF-augmented-on-policy-70b 0 0 0

davisrbr / jailbreakbench-goal-embeddings-artifacts 0 0 0 2

bogdanminko / wildguardmix-cleaned 0 0 0 2

thunder-research-group / SNU_Ko-IFEval gated 0 0 1

aisingapore / Instruction-Following-IFEval gated 0 0 2

InazakiSakuya / llama-guard-harmful 0 0 0 2

projecte-aina / IFEval_ca 0 0 0 7

jkazdan / harmbench-gpt-4o 0 0 0

david9dragon9 / cvalues_samples 0 0 0

david9dragon9 / xs_rl_data 0 0 0 1

gaydmi / IFEval_tt gated 0 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI 0 0 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI-hard-no 0 0 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI-AMD 0 0 0 1

NoorNizar / HarmBench-Test-Standard 0 0 0 2

AlignmentResearch / XSTest 0 0 0 1

jerome-white / leaderboard-documents-ifeval 0 0 0

jimim / IFEval_subset100 0 0 0

Ayush-Singh / IFEval_ours_llama 0 0 0

Ayush-Singh / IFEval_self_llama 0 0 0 1

Ayush-Singh / IFEval_reward_llama 0 0 0

YinkaiW / harmbench-dataset 0 0 0 2

xchen16 / wjb_responses 0 0 0 1

ata990 / HarmBench_Base64 0 0 0 2

hbseong / HarmAug_generated_dataset 0 0 0 2

AlignmentResearch / HarmBench 0 0 0 1

rasdani / smol-RLVR-IFeval 0 0 0 1

CocoRoF / IFEval-distill 0 0 0 1

llamastack / IfEval 0 0 0 1

linggm / processed-ifeval 0 0 0

innodatalabs / rt3-wildguardmix 0 0 0

Jianshu001 / ifeval_dpsk 0 0 0 1

BrianGu / harmbench_behaviors 0 0 0 1

INSAIT-Institute / IFEval-bg 0 0 0 1

lucweber / ifeval-like-data__subsampled 0 0 0

pxyyy / ifeval-like-data 0 0 0 1

pxyyy / ifeval-like-data-20k 0 0 0

pxyyy / Llama-3.1-8B-autoif-20k-1-1e-5-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top2000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top15000-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top2000-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top15000-by-mp-for-ifeval-correct-200 0 0 0

agentlans / drill 0 0 0 1

pxyyy / Llama-3.1-8B-autoif-20k-1-1e-5-ifeval-correct-200-with-doc-id 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-1e-4-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top2000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top15000-by-mp-1e-4-for-ifeval-correct-200 0 0 0

INSAIT-Institute / ifeval_ukr 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-1e-3-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top2000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top15000-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-3e-4-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top1500-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top2000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top15000-by-mp-3e-4-for-ifeval-correct-200 0 0 0

xsample / tulu-3-car-50k 0 0 0

xsample / tulu-3-ifd-50k 0 0 0 1

xsample / tulu-3-instag-50k 0 0 0

xsample / tulu-3-qdit-50k 0 0 0

xsample / tulu-3-random-50k 0 0 0

xsample / tulu-3-zip-50k 0 0 0

pxyyy / autoif-20k-removed-top500-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top1500-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top2000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top5000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top10000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top15000-by-mp-1e-3-thre95-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top12500-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top17500-by-mp-1e-3-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top12500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top17500-by-logix-for-ifeval-correct-200 0 0 0

pxyyy / autoif-20k-removed-top7500-by-mp-1e-3-for-ifeval-correct-200 0 0 0 1

pxyyy / autoif-20k-removed-top7500-by-logix-for-ifeval-correct-200 0 0 0

ai2-adapt-dev / tulu-3-thinker-classified-no_math-ifeval-ref-based-45k 0 0 0 1

Yeganeh / IFEval-labeled 0 0 0

AlignmentResearch / WildGuardTest 0 0 0 1

ai2-adapt-dev / tulu_3_rewritten_100k-v2-ifeval 0 0 0 1

GenRM / magpie-ultra-v1.0-argilla 0 0 0

Beijing-AISI / panda-bench 0 0 0 2

ai2-adapt-dev / tulu_3_rewritten_100k-v3-ifeval-no_math 0 0 0 1

faezeb / tulu_3_rewritten_100k-v3-ifeval-general-no-ref 0 0 0 1

tungluuai / wildguard-multiple-questions gated 0 0 1

rasdani / ifeval-genesys 0 0 0 1

rasdani / ifeval-genesys-verified 0 0 0

rasdani / ifeval-genesys-debug 0 0 0

jdineen / or-bench 0 0 0 1

ai2-adapt-dev / tulu_3_rewritten_100k-v2-ifeval-replaced-math 0 0 0 1

hamishivi / RLVR-IFeval 0 0 0

BrachioLab / jailbreak_harmbench50 0 0 0 2

faezeb / RLVR-IFeval 0 0 0 1

iagoalves / wildguardmix_Qwen3-8B 0 0 0 1

iagoalves / wildguardmix_Qwen3-8B_jb 0 0 0

iagoalves / wildguardmix_Qwen3-8B_jb_rs 0 0 0

Lv111 / harmbench_behaviors_text_all 0 0 0

MCINext / persian-ifeval 0 0 0

iagoalves / wildguardmix_Qwen3-8B_jb_full 0 0 0

iagoalves / wildguardmix_Qwen3-8B_jb_guardrail 0 0 0

faezeb / RLVR-IFeval-no-math 0 0 0

iagoalves / wildguardmix_Qwen3-8B_notjb_rs 0 0 0

iagoalves / wildguardmix_Qwen3-8B_full_rs 0 0 0

iagoalves / wildguardmix_Qwen3-8B_full_rsgr 0 0 0

iagoalves / wildguardmix_eval 0 0 0 1

522H0134-NguyenNhatHuy / test_1K_prompts_vie 0 0 0

justus27 / ifeval-validation-test 0 0 0

kowndinya23 / ifeval_prompts 0 0 0

tokyotech-llm / M-IFEval-Ja 0 0 0

MatanBT / gcg-evaluated-data 0 0 0 2

iagoalves / wildguardmix_train_puro 0 0 0 1

Impel / Jailbreak_Benchmark 0 0 0

JW17 / WildguardMix-non-ad-45k-Q25-1.5B-SFT 0 0 0 1

JW17 / WildguardMix-non-ad-45k-Q3-1.7B 0 0 0

JW17 / WildguardMix-Q3-1.7B-labeled 0 0 0

JW17 / WildguardMix-Q25-SFT-labeled 0 0 0

VGraf / ifeval_mt 0 0 0 1

japan-ai-official / jifeval 0 0 0 1

Compumacy / HarmBench-Test-Standard 0 0 0

BRlkl / WildGuardTest-pt 0 0 0 1

BRlkl / XSTest-pt 0 0 0 1

Sakonii / task-over-refusal-dataset 0 0 0 1

saumyamalik / wildguardmix-r1-v2-format-filtered 0 0 0

justus27 / prime-rl-ifeval 0 0 0

TAUR-dev / IFEval__qwen__bon__single 0 0 0

Thanmay / IFEval 0 0 0 1

NoahShen / harmbench-llama3.1-8b-inst-completions 0 0 0

NoahShen / xstest-llama3.1-8b-inst-completions 0 0 0 1

davidkim205 / ko-ifeval gated 0 0

AISafety-Student / reasoning-safety-behaviours gated 0 0

mariagrandury / harmbench_copyright_classifier_hashes 0 0 0

alexanderstern / harmbench_behaviors 0 0 0

alexanderstern / harmbench 0 0 0 1

dnouv / prompt_guardrail_eval 0 0 0 1

Thanmay / IFEval-verified 0 0 0 1

ilsp / Jailbreak-StrongReject-el 0 0 0 1

sa7270 / RLVR-IFEVAL-Filtered 0 0 0 1

ddedaniel02 / bert-jailbreak-classifier 0 0 0

purpcode / XSTest 0 0 0

purpcode / malevent-jailbreak-eval-oss 0 0 0 1

purpcode / jailbreak-eval-harm-prompts 0 0 0

purpcode / malevent-jailbreak-eval-artprompt 0 0 0

tartuNLP / ifeval_et 0 0 0 1

oneonlee / wildguardmix-train 0 0 0 1

DhruvTre / h2-harmbench-twins 0 0 0 1

jackzhang / JBDistill-Bench 0 0 0 1

BRlkl / wildguardmix-pt 0 0 0 1

DhruvTre / jailbreakbench-paraphrase-2025-08 0 0 0 2

saumyamalik / tulu-v3-9-synthetic-finalresp-wildguardmixtrain-decontaminated-50k-unused-qwq 0 0 0

saumyamalik / tulu-v3-9-synthetic-finalresp-wildguardmixtrain-decontaminated-50k-unused-gemma3 0 0 0

zsqzz / wildguard-adversarial-harmful-3k 0 0 0

evats / StrongReject_el 0 0 0 1

zsqzz / wildguard-3k-formal 0 0 0

Oluwadara / JBB-Behaviors 0 0 0 1

NoahShen / id-0002-harmbench 0 0 0

NoahShen / id-0002-xstest-completions 0 0 0

NoahShen / id-0010-harmbench 0 0 0

NoahShen / id-0010-jailbreakbench 0 0 0

NoahShen / id-0003-xstest 0 0 0

NoahShen / id-0004-xstest 0 0 0

NoahShen / id-0010-xstest 0 0 0

NoahShen / id-0011-xstest 0 0 0 1

NoahShen / id-0009-xstest 0 0 0

amang1802 / pac-bench-100pct-unaligned 0 0 0

manan-u / ifeval-translated 0 0 0 1

NoahShen / id-0001-beear-harmbench 0 0 0

NoahShen / id-0001-beear-xstest 0 0 0

friendshipkim / IF-Datasets-Tulu-IFEval 0 0 0 1

NoahShen / id-0001-baseline-beear-harmbench 0 0 0 1

NoahShen / id-0010-beear-xstest 0 0 0

NoahShen / id-0009-beear-xstest 0 0 0

NoahShen / id-0011-beear-xstest 0 0 0

NoahShen / id-0010-unlearn-harmbench 0 0 0

NoahShen / id-0010-unlearn-xstest 0 0 0

NoahShen / id-0009-unlearn-xstest 0 0 0

NoahShen / id-0011-unlearn-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-xstest 0 0 0

NoahShen / id-0009-gcg-unlearn-xstest 0 0 0

NoahShen / id-0010-gcg-unlearn-harmbench 0 0 0

NoahShen / id-0010-gcg-unlearn-xstest 0 0 0

NoahShen / id-0011-gcg-unlearn-xstest 0 0 0

amang1802 / pac-bench-100pct-unaligned_rephrased 0 0 0

val-bench / calibration-unaligned-0 0 0 0

NikolaiSkripko / IFEval-FC 0 0 0 2

Thanmay / IFEval-sangraha-1 gated 0 0 1

Thanmay / IFEval-sangraha-3 gated 0 0

Thanmay / IFEval-sangraha-2 gated 0 0 1

Thanmay / IFEval-annotations 0 0 0

mikasenghaas / allenai-ifeval_gpt-5-mini_n5_r3 0 0 0 1

Jazhyc / wildguardmix-annotations 0 0 0 1

anushakamath / test_i 0 0 0

jaxon3062 / IFEval-gemma3-chat 0 0 0 1

mvrcii / safety-refusals 0 0 0

mvrcii / safety-harmful 0 0 0

boczkakaroly / trilingual-cultural-bias-redteaming-benchmark 0 0 0 1

shiv96 / HarmBench-standard-llama_3.2-1B-it-activation-pca-steered 0 0 0

shiv96 / HarmBench-standard-llama_3.2-1B-it-kpca-steered 0 0 0 1

Seanie-lee / ThinkSafe-R1-Distill-1.5B-v3 0 0 0

sungyub / ifeval-rlvr-verl 0 0 0

sungyub / if-verl-unified 0 0 0

puwaer / do_not_answer_en_response 0 0 0 1

puwaer / do_not_answer_jp_response 0 0 0

puwaer / do_not_answer_zh_response 0 0 0

Genteki / xstest_tiny 0 0 0

tartuNLP / ifeval_en 0 0 0 1

jbrin / or-bench-balanced 0 0 0 1

Neelectric / wildguardmix_Llama-3.1-8B-Instruct_4096toks gated 0 0 1

FlameF0X / Safety_Alignment_Benchmark gated 0 0 1

richardyoung / tempest-replication 0 0 0 1

denis1699 / jailbreak-benchmark-dataset-benign-advbench 0 0 0 1

denis1699 / jailbreak-benchmark-dataset-benign-gsm8k 0 0 0 1

sentinelseed / sentinel-evaluations 0 0 0 1

TheFactoryX / edition_1314_google-IFEval-readymade 0 0 0

aiymka / JailbreakBench 0 0 0 1

Rendra86318 / IFEval 0 0 0 1

zhqwqwq / SFT_eval 0 0 0 1

agentlans / refusal-classifier-data 0 0 0 1

Aurther-Nadeem / introspect-ai-benchmark 0 0 0 1

pant0m / harmbench_classifier_train 0 0 0 1

CatBarks / harmbench_humanjailbreaks_llama2_strictjudge_success_only 0 0 0 1

YuminChoi / ThinkSafe-0.6B-n4-filtered-LlamaGuard 0 0 0

YuminChoi / ThinkSafe-1.7B-n4-filtered-LlamaGuard 0 0 0

YuminChoi / ThinkSafe-4B-n4-filtered-LlamaGuard 0 0 0

YuminChoi / ThinkSafe-8B-n4-filtered-LlamaGuard 0 0 0 1

rana-shahroz / ifeval-cot-llama3.3-70B 0 0 0 1

tonychenxyz / ifeval 0 0 0 1

zyL2003 / JBB-Behaviors 0 0 0 1

lamhots / BatakJailbreakBench 0 0 0 1

YuminChoi / STAR-41K-llama-guard 0 0 0

haralab-uec / steering-bench-ja 0 0 0 2

jamesbowen / JBB-Behaviors 0 0 0 1

Brantliu / FalseReject 0 0 0

Seanie-lee / ThinkSafe-0.6B-WildGuard 0 0 0 1

Seanie-lee / ThinkSafe-1.7B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-R1-Distill-8B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-4B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-8B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-R1-Distill-1.5B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-R1-Distill-1.5B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-R1-Distill-8B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 0 1

icml2026-7516 / ThinkSafe-Qwen3-4B-WildGuard 0 0 0

icml2026-7516 / ThinkSafe-Qwen3-8B-WildGuard 0 0 0

Seanie-lee / ThinkSafe-R1-Distill-7B-WildGuard 0 0 0

ramendik / kimify-ifeval-like 0 0 0 1

icml2026-7516 / ThinkSafe-R1-Distill-7B-WildGuard 0 0 0 1

Neelectric / Replay_0.01.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

Neelectric / Replay_0.02.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

Neelectric / Replay_0.03.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

Neelectric / Replay_0.04.MoT_science.wildguardmix.Llama3_4096toks 0 0 0

Neelectric / Replay_0.05.MoT_science.wildguardmix.Llama3_4096toks 0 0 0 1

Neelectric / wildguardmix_reasoning_Llama-3.1-8B-Instruct_4096toks 0 0 0 1

jcanode / safeguard-prompt-injection 0 0 0

Seanie-lee / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 0 1

Seanie-lee / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 0 1

Seanie-lee / ThinkSafe-Qwen3-4B-WildGuard 0 0 0 2

Seanie-lee / ThinkSafe-Qwen3-8B-WildGuard 0 0 0 1

theblackcat102 / ifeval_m 0 0 0

Neelectric / wildguardmix_reasoning_Llama-3.1-8B-Instruct_4096toks_refusals_only 0 0 0 1

Neelectric / wildguardmix_Llama-3.1-8B-Instruct_4096toks_refusals_only 0 0 0

Neelectric / Replay_0.1.MoT_science.wildguardmix.Llama3_4096toks 0 0 0 1

tytodd / or_bench-qwen3-vl-32b-l16 0 0 0

ClarusC64 / ai-constitutional-helpfulness-drift-suite-v0.1 0 0 0

xuanduy04 / wildguardmix_with_guard_label gated 0 0

acartag7 / gap-benchmark 0 0 0 1

jmk9494 / moderation-bias-benchmark 0 0 0 1

mohdelgaar / ifeval_rlvr 0 0 0 1

Yehiaa06 / harmbench-evaluation 0 0 0

ceselder / concept-cot-corpus-mini 0 0 0

ceselder / concept-cot-qa-mini 0 0 0

eve6352 / RobloxGuard-Eval 0 0 0 1

IanSteenstra / AI-Psychotherapy-Eval 0 0 0 1

haritzpuerto / ifeval-lrm 0 0 0 1

eazzy / wildguard_for_distil 0 0 0 1

LLMSafety / HarmBench 0 0 0 1

Faruna01 / lsr-benchmark 0 0 0 1

Lhmisme / vata-benchmark 0 0 0

johnzqlu / grm-30b-a3b-math-ifeval-global-829_soft_judgments 0 0 0

alpha-one-index / ai-red-teaming-index 0 0 0

ontocord / Dolci-Think-RL-7B-decontaminated gated 0 0

Hhhhhl / JBB-Behaviors 0 0 0 1

jbross-ibm-research / M-IFEval 0 0 0 1

saiteki-kai / XSTest-IT 0 0 0

HiTZ / ifeval_eu 0 0 0 1

HiTZ / ifeval_gl 0 0 0

taowangcheng / Original-IFEval 0 0 0

taowangcheng / IFEval 0 0 0

shuyilin / JO-Bench 0 0 0 1

FearedFusionX / RobloxGuard-Eval 0 0 0 1

kyubeen / temp 0 0 0

minyichen / IF-R1 gated 0 0 1

timtkddn / Guardrail-eval-xstest-beavertail-30k-en-ko gated 0 0 1

HAC-Studios-Org / prompted-hearts-clinical-eval 0 0 0 1

safety-aya / harmbench-portuguese 0 0 0 1

safety-aya / wildguardtest-portuguese 0 0 0

safety-aya / xstest-portuguese 0 0 0

wvnvwn / wildguard 0 0 0

wvnvwn / wildguardmix 0 0 0

Tokyomonster / JBB-Behaviors 0 0 0 1

NoahShen / id-0001-unlearn-seed-42-jailbreakbench 0 0 0

NoahShen / id-0001-unlearn-seed-42-xstest 0 0 0

NoahShen / id-0001-unlearn-seed-519-jailbreakbench 0 0 0

NoahShen / id-0001-unlearn-seed-519-xstest 0 0 0

NoahShen / id-0001-unlearn-seed-519-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-42-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-1024-harmbench 0 0 0

NoahShen / id-0001-unlearn-seed-1024-xstest 0 0 0

NoahShen / id-0001-unlearn-seed-2048-xstest 0 0 0

NoahShen / id-0001-unlearn-seed-2048-harmbench 0 0 0

NoahShen / id-0001-beear-seed-42-harmbench 0 0 0

NoahShen / id-0001-beear-seed-42-xstest 0 0 0

NoahShen / id-0001-beear-seed-519-xstest 0 0 0

NoahShen / id-0001-beear-seed-1024-xstest 0 0 0

NoahShen / id-0001-beear-seed-519-harmbench 0 0 0

NoahShen / id-0001-beear-seed-2048-xstest 0 0 0

NoahShen / id-0001-beear-seed-1024-harmbench 0 0 0

NoahShen / id-0001-beear-seed-2048-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-42-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-42-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-519-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-519-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-1024-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-2048-xstest 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-1024-harmbench 0 0 0

NoahShen / id-0001-gcg-unlearn-seed-2048-harmbench 0 0 0

silmarilsecurity / firewall-eval-v1 gated 0 0 1

wvnvwn / LOCA-wildguard-offline 0 0 0

wvnvwn / LOCA-wildguard 0 0 0

wvnvwn / LOCA-wildguardmix 0 0 0

adityabagayatkar / JBB-Behaviors 0 0 0 1

Micdejc / cymultenset 0 0 0 1

intherejeet / agentic-rag-redteam-bench gated 0 0 1

davidquicast / sorry-bench-202503-multilingual-backup 0 0 0 1

HJH2CMD / lrm-safety-eval 0 0 0 1

boolishs / xstest-v2-copy 0 0 0 1

kairawal / MultiLingual-SorryBench 0 0 0 1

tytodd / qwen3.5-4b-or_bench_toxic 0 0 0

dmody1 / wildguardmix-refusal-generations 0 0 0 1

clulab / LEG-1.0-wildguardmix-explanations gated 0 0

agentic-moral-alignment / gt-harmbench-eval 0 0 0 1

Turbs / xprmt-tiny-aya-global-advbench-judge-eval 0 0 0

PowCal / HarmBench100 0 0 0 1

agentic-moral-alignment / gt-harmbench-deont 0 0 0

jang1563 / bio-overrefusal-v0.1 0 0 0 1

tytodd / qwen3.5-2b-v2-smoke 0 0 0

tytodd / gpt-oss-20b-v2-no-gepa 0 0 0

imadreamerboy / constitutional-safety-classifier-data 0 0 0 1

richardyoung / code-safety-benchmark-results gated 0 0

zzyppp / ifeval 0 0 0 2

abhi14 / rm-axis-blindness-experiment 0 0 0

stablegradients / IF-Datasets-Tulu-IFEval 0 0 0 1

zypho / ifeval 0 0 0

CapyStudios / capydataset 0 0 0 1

chloeclerc17 / English_French_safety_code-mixing_dataset 0 0 0 1

choiqs / IF-Datasets-Tulu-IFEval 0 0 0 1

Jinesis / gt-harmbench 0 0 0 1

ZachW / gemma-3-27b-it_ifeval 0 0 0

ZachW / gemma-4-31b-it_ifeval 0 0 0

ZachW / gpt-oss-20b_ifeval 0 0 0

ZachW / nanbeige4-3b-thinking-2511_ifeval 0 0 0

ZachW / olmo-3-7b-think_ifeval 0 0 0

ZachW / qwen3-30b-a3b_ifeval 0 0 0

ZachW / qwen3-32b_ifeval 0 0 0

ZachW / qwen3-8b_ifeval 0 0 0

shalanova / benchmark-3-arabic-gt 0 0 0

shalanova / benchmark-4-chinese-gt 0 0 0

shalanova / benchmark-3-russian-gt 0 0 0

shalanova / benchmark-3-chinese-gt 0 0 0

shalanova / benchmark-4-russian-gt 0 0 0

shalanova / benchmark-4-arabic-gt 0 0 0

shalanova / benchmark-3-arabic-m2m 0 0 0

shalanova / benchmark-3-chinese-m2m 0 0 0

shalanova / benchmark-3-russian-m2m 0 0 0

shalanova / benchmark-4-arabic-m2m 0 0 0

shalanova / benchmark-4-chinese-m2m 0 0 0

shalanova / benchmark-4-russian-m2m 0 0 0

UCSC-VLAA / CIK-Bench 0 0 0 1

Jocana / herald-logits 0 0 0 1

samuelsimko / rr-circuit-breakers-attack-completions gated 0 0 1

bridge-benchmark-2026 / BRIDGE 0 0 0 1

neurips2026-10977 / ThinkSafe-R1-Distill-1.5B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-R1-Distill-7B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-R1-Distill-8B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 0

neurips2026-10977 / ThinkSafe-Qwen3-4B-WildGuard 0 0 0 1

neurips2026-10977 / ThinkSafe-Qwen3-8B-WildGuard 0 0 0

sleightbench / SLEIGHT-Bench gated 0 0

abliterationai / authorized-cyber-policy-eval 0 0 0

abliterationai / governed-ai-actions-bench 0 0 0

abliterationai / synthetic-data-provenance-eval 0 0 0 1

abliterationaiorg / governed-ai-actions-bench 0 0 0

abliterationaiorg / synthetic-data-provenance-eval 0 0 0

AItherapist / DBE-Bench gated 0 0

rosafety-bench / rosafety-bench 0 0 0

Panuthep / wildguardmix_2k 0 0 0 1

jkminder / model-raising-pb-300k-3c-sft 0 0 0 1

jkminder / model-raising-pb-100k-3c-mt-sft 0 0 0 1

VityaVitalich / or-bench-1k 0 0 0 1

jacobmorrison / cse-579-ifeval-rl 0 0 0 1

cs-552-2026-vibe-trainers / mcq_safety 0 0 0 1

Thanmay / IFEval-pipe 0 0 0 1

jkminder / xstest-overrefusal 0 0 0

jkminder / or-bench-1k-overrefusal 0 0 0 1

AbdullahRashid / mode-probe-bench 0 0 0

massaindustries / dataset-A-routing 0 0 0 1

de-Rodrigo / ifeval-format-clean 0 0 0 1

anonymous-review-anonymous / HarmMetric_Eval 0 0 0 1

de-Rodrigo / ifeval-decontaminate 0 0 0 1

2etatg / abliteration-eval 0 0 0

LLMSafety / XSTest 0 0 0

m-zest / trojanspec-bench 0 0 0 1

khaledyusuf44 / somalibench-v0 0 0 0 1

szzh12 / JBB-Behaviors 0 0 0 1

zzha6204 / Health-ORSC-Bench 0 0 0 1

zdxnlp / SaladBench 0 0 0 1

EliasHossain / ptc-benchmark 0 0 0 1

balabalab / JBB-Behaviors 0 0 0 1

hirundo-io / or-bench 0 0 0 1

anonymous-emnlp-2026-ejt / EJT_Benchmark gated 0 0 1

mm-eval / MM-IFEval 0 0 0

kasadin / alignment-4type 0 0 0 1

Jaemo3123 / JBB-Behaviors 0 0 0 1

sci4ai / Qwen3.5-27B-Abliterated-HarmBench-Refusals 0 0 0 1

hirundo-io / XSTest 0 0 0

hivetrace / harm_bench_completions gated 0 0 1

schaeff / CIAware-Bench-Data gated 0 0

Ethosoft / Mycelium-Eval-Artifact 0 0 0

cs-552-2026-databand / Safety_Dataset_MCQ 0 0 0 1

LLMSafety / or-bench 0 0 0

cs-552-2026-qwenlifegivesyoulemons / safety_sft_data 0 0 0

cs-552-2026-qwenlifegivesyoulemons / safety_aegis_mcqa 0 0 0 1

moon12321 / WildGuardTest 0 0 0 1

PhilippBlue / mcq_safety_boxed 0 0 0 1

luna-pragma-2026 / Decontam-Bench 0 0 0 1

Nid4l / X-Guard-Bench gated 0 0

stindardlogic / refusal-overrefusal-50k 0 0 0 1

Jnx03 / kanitakorn-deepseek-v46-v45-if-mt-hotpot-bridge-mix 0 0 0

solanaclawd / solana-clawd-eval 0 0 0 1

FinancialSupport / IFEval-max-tokens 0 0 0 1

kurakurai / IFEval-FR 0 0 0 1

sincpp / AgentInjectionBench 0 0 0

kevin-giskard / XSTest 0 0 0 1

lego17440 / guardrail-input-slm-data 0 0 0 1

UltraMar1ne / residual-streams-20260618 gated 0 0 1

giskardai / do-not-answer-scenarios 0 0 0 1

uninhibited-scholar / defensive-refusal-bench-zh 0 0 0 1

yavuz-ai / seas-harden-only 0 0 0

yavuz-ai / seas-coevolve 0 0 0 1

yavuz-ai / seas-harden-ctx 0 0 0

PowerBench / PowerBench-Multilingual 0 0 0 1

PowerBench / PowerBench-Geopolitics 0 0 0

anna-sarvam / indic-safety-eval 0 0 0 1

zanellac / m-ifeval-partages 0 0 0

hirundo-io / bloom-over-refusal-free-text 0 0 0 1

giskardai / harmbench-scenarios 0 0 0 1

SabrinaSadiekh / responses-and-asr-labels-small-models 0 0 0 1

ariffazil / a2b-eval-results 0 0 0 1

Aniskhan29 / harmbench_behaviors 0 0 0

lefft / needleif-bench 0 0 0 1

Tomk187 / LongHarmBench gated 0 0 1

jamesdborin / IFEval-prompt-only 0 0 0

SEBK4C / gemma4-serving-bench-data 0 0 0 1

briefloop / eval-packs 0 0 0

astroware / halo-guard-bench 0 0 0 1

o22y / unrelabel-poison-benchmark 0 0 0

devicemark / results 0 0 0 1

budecosystem / IFEval 0 0 0

kepom / JBB-Behaviors 0 0 0

isisyangyang / judgeflip-judge-audit 0 0 0

SMahjuba / A-Tale-of-Two-Nations-AI-Benchmark 0 0 0

sripad17 / harm-probe-train-en gated 0 0

franktheglock / temp-guard-data-20260722 0 0 0

lamduong / human-ai-impact-bench-scenarios 0 0 0 1

ividal / harmbench-copyright-hashes 0 0 0

fevziegeyurtsevenler / guardrail-arena 0 0 0

fevziegeyurtsevenler / turkish-over-refusal-set 0 0 0

quanshu01 / SPIKE-Bench gated 0 0

jerogo / or-bench 0 0 0 1

Magurofg / ifeval-pt 0 0 0 1

invarra / jailbreak-control-benchmark-v1-metadata 0 0 0

Cyber-security-final-project / Evaluation_of_OpenSource_Models_for_PDF_Injection_Recognition 0 0 0 1

fevziegeyurtsevenler / PolyGuardBench 0 0 0 1

digitable-lol / digit-eval-tasks 0 0 0 1

rustem17 / em-code-subliminal-transfer-evals 0 0 0 1

saintsauce / uniar-steering-eval 0 0 0 1

APTO-001 / APTO-or-bench-ja gated 0 0

shrugging-shoulders / ifeval_multilang 0 0 0 1

wflying / instruction-following-rl-66k 0 0 0

loaiabdalslam / RedTeam-LLM-Benchmark-v2 0 0 0

loaiabdalslam / RedTeam-LLM-Benchmark-v2-Results 0 0 0

hkab / vi-ifeval 0 0 0 1

MatanBT / JBRL-robustness-eval gated 0 0 1

lilyzhng / c-guard 0 0 0 1

kosiasuzu / tulu-postraining-data 0 0 0 1

Ishowbackup / JBB-Behaviors 0 0 0 1

open-athena / glm52-datagen-r11-109-inverse-ifeval-v2-traces 0 0 0

atharv146 / college-advice-ai-safety-eval-prompts 0 0 0

hungbenjamin402 / IF-multi-constraints-upto5-LFM2.5-prompts 0 0 0

hungbenjamin402 / IF-RLVR-IFEval-prompts-LFM2.5 0 0 0

hungbenjamin402 / Nemotron-RL-IF-Gym-LFM2.5-prompts 0 0 0

groovinandmovin / JBB-Behaviors 0 0 0 1

jensjepsen / danish-if-grpo-combined-v3 0 0 0 1

rlundqvist / vea-generalization-benchmark 0 0 0

Siye01 / LLM-Fusion-Test 0 0 0 1

rlundqvist / ifeval-obf-rl-run 0 0 0 1

rlundqvist / ifeval-obf-rl-run-leak 0 0 0 1

adityanawal23 / HarmURLBench gated 0 0 1

chibifire / harmful-prompts-bench 0 0 0

lgoyal / open-refusal-steering 0 0 0

hamiejuice / qwen3.8-27b-uncensored-dflash2-m4-pro-benchmark 0 0 0 1

frangelbarrera / ling-3-flash-evaluation 0 0 0 1

shrugging-shoulders / inverse_ifeval_multilang 0 0 0 1

open-athena / nemotron-gym-cfbench 0 0 0

open-athena / nemotron-gym-inverse-ifeval 0 0 0

open-athena / nemotron-gym-sysbench 0 0 0

eda-bench-neurips-2026 / eda-bench-public-provenance 0 0 0 1

ecairol / simpleqa-eval-qwen3.8-27b-obliterated 0 0 0 2

abeeranajam31 / CLSG-Benchmark 0 0 0 1

leexiaoyi-lab / HDHC-Benchmark 0 0 0 1

phionyx / airep-embedded-evaluation-profile 0 0 0 2

dungnv / trail-14b-activations 0 0 0

TuanAnhHoangBui / qwen3-safety-guard-scaled-mixtures-private 0 0 0

CounterSteer / overrefusal-data 0 0 0

noahrossi / heretic-completions 0 0 0 1

CounterSteer / overrefusal-artifacts 0 0 0 1

GOSHUNCLE / tc-slang-guardrail-benchmark gated 0 0

shahoismael / prompt-quality-vs-response-compliance 0 0 0 1

AriaAICompany / or-bench-problems 0 0 0

Sangsang / ThinkSafe-Qwen3-8B-ICL-data 0 0 0

Sangsang / ThinkSafe-Qwen3-8B-Activation-data 0 0 0 1

Sangsang / ThinkSafe-DeepSeek-R1-Distill-Llama-8B-ICL-data 0 0 0

Sangsang / ThinkSafe-DeepSeek-R1-Distill-Llama-8B-Activation-data 0 0 0

emgena / automated_redteaming_adversarial_jailbreak_eval_teaser 0 0 0

quickium / awall-eval-benchmark 0 0 0 1

muhammadhunzla971-sys / roman-urdu-blindspot-eval 0 0 0

mmse-bench-anon / msb 0 0 0 1

aakash-agarwal / Indic-Jailbreak-Bench gated 0 0 1

hanxunh / VEX-Bench 0 0 0 1

SUAN-PO / OR-Bench 0 0 0

SUAN-PO / HarmBench 0 0 0

false-facts-finetuning / lr-sweep-evals 0 0 0

Teraformer-limited / evalboard-devicemark-raw 0 0 0

Arushhh / phase2-drafter-swap 0 0 0

Neelectric / wildguardmix_OLMo-2-1124-7B-Instruct_4096toks_refusals_only 0 0 0

LANA-research / LANA-IF-Qwen3.8-Distill gated 0 0

farbodghasemlu / bonsai-27b-q1-abliteration-benchmark 0 0 0

hirundo-io / secfid-agentic-refusal-qa 0 0 0

Center-Of-Advanced-Software-Technologies / IFEval-Multi-IF-hy 0 0 0 1

karthik789338 / FailureAtlas-IS 0 0 0

Crusadersk / quantfit-reference-reports 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_prefill_flip_n52_chat 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_prefill_flip_n52_plain 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.strict_compliance_refusal 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_ladder 0 0 0

siddharthmb / 2026.TA.experiments.refusal_tracing.refusal_prefill_flip_n3 0 0 0

iconseven / trustboundary-redteam-benchmark 0 0 0

ngila-ai / ngila-benchmarks-ccby-v0.2 0 0 0

ngila-ai / ngila-benchmarks-mit-v0.2 0 0 0

ngila-ai / ngila-do-not-answer-v0.2 0 0 0 1

ValeEverheart / vale-governance-eval 0 0 0 1

Hastika06 / dualr-benchmark 0 just added 0 1

AWKLjdnasjkfbsakj / RLVR-IFeval 0 just added 0

alfaplus / backdoor-detect-bench-v1 0 just added 0 1

submarat / mats11-mix4-readout2x 0 just added 0 1

RayNene / qwen3-multilingual-safety-eval 0 just added 0 1

Infatoshi / DeepSeek-V4.1-Flash-1017-local-evaluation 0 just added 0

NoahIsARider / RecRedTeam-ASI-Bench 0 just added 0

fatima2442 / qwen25-multilingual-safety-eval 0 just added 0

amyrmahdy / decima-guard-bench 0 just added 0

danielrosehill / The-PC-Bench 0 just added 0

allenai / wildguardmix gated 0 2

google / IFEval 0 0 7

Horizon-Labs / prompt-injection-eval-suite 0 0 1

walledai / XSTest gated 0 1

argilla / magpie-ultra-v1.0 0 0 4

MultiverseComputingCAI / llm-refusal-evaluation 0 0 1

Necent / llm-jailbreak-prompt-injection-dataset gated 0 1

votal-ai / ai-redteaming-safety-model 0 0 1

masahiroid / JBB-Behaviors-ja 0 0 1

andersonbcdefg / redteaming_eval_pairwise 0 0 1

ethz-spylab / competition_eval_dataset 0 0 1

yuchenlin / just-eval-redteam 0 0 1

ethz-spylab / harmless-eval-SUDO gated 0 1

LLM-Tuning-Safety / HEx-PHI gated 0 1

harpreetsahota / IFEval_Experiments 0 0 1

natolambert / xstest-v2-copy 0 0 1

OpenSafetyLab / Salad-Data 0 0 2

hf-100 / llama-guard 0 0 1

mcj311 / saladbench_data 0 0 1

markush1 / LLM-Jailbreak-Classifier gated 0 2

ConiferLM / Conifer 0 0 2

xiaodongguaAIGC / CValues 0 0 1

yimingzhang / hh-rlhf-safety 0 0 1

markush1 / LLM-Jailbreak-Classifier-Large gated 0 4

bench-llm / or-bench 0 0 2

onepaneai / gpt-harmful-prompts-after-guardrails-evaluation 0 0 2

onepaneai / falcon-harmful-prompts-before-guardrail-evaluation 0 0 2

thobauma / harmless-eval-SUDO-random-double-space 0 0 2

onepaneai / falocn-harmful-prompts-after-guardrails-evaluation 0 0 1

onepaneai / harmful-prompts-gpt-before-guardrail-evaluation 0 0 1

onepaneai / harmful-prompts-falcon-after-guardrail-evaluation 0 0 1

onepaneai / harmful-prompts-gpt-after-guardrail-evaluation 0 0 1

onepaneai / harmful-prompts-gemini-pro-after-guardrail-evaluation 0 0 1

onepaneai / harmful-prompts-multillm-after-guardrail-evaluation 0 0 1

onepaneai / harmful-prompts-multillm-before-guardrail-evaluation 0 0 1

onepaneai / harmful-prompts-before-guardrail-mixtral-evaluation-latest 0 0 1

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-latest 0 0 1

onepaneai / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 2

onepaneai / harmful-prompts-after-guardrail-mixtral-evaluation-new-200 0 0 1

amilmshaji / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 3

amilmshaji / harmful-prompts-after-guardrail-mixtral-evaluation-latest 0 0 1

sajosam / harmful-prompts-multillm-after-guardrail-evaluation-new 0 0 2

IS2Lab / S-Eval 0 0 2

dynamoai / safe_eval 0 0 1

thobauma / fixed-harmless-eval-SUDO-random-double-space 0 0 1

shane131214 / harmless-eval-SUDO 0 0 1

Therax / llama_guard 0 0 1

bench-llms / or-bench 0 0 1

walledai / StrongREJECT gated 0 1

walledai / JailbreakBench 0 0 1

walledai / HarmBench gated 0 4

randxie-boson / ifeval-data 0 0 1

shane131214 / harmless-eval-SUDO_with_defense 0 0 1

shane131214 / harmless-eval-SUDO_enhence 0 0 2

thobauma / fixed-harmless-eval-SUDO-random 0 0 1

harpreetsahota / llama3_1-405B-on-IFEval 0 0 2

walledai / SGXSTest gated 0 1

longphann / harmbench_classifier_train 0 0 1

jzhang86 / corrected_ifeval 0 0 1

bezir / IFEval_tr 0 0 4

Nutanix / allenai-wildguardmix-processed gated 0 1

AmanPriyanshu / FRACTURED-SORRY-Bench-Automated-Multishot-Jailbreak gated 0 1

ToxicityPrompts / wildguard-train 0 0 1

heegyu / XSTest_ko 0 0 2

heegyu / xstest-response-ko 0 0 1

PhillipGuo / mistral_7b_prompt_completion_dataset 0 0 2

mjalg / IFEval_tr 0 0 2

AnonHB / HarmAug_generated_dataset 0 0 3

gabrielmbmb / ifeval-like-data 0 0 2

argilla / ifeval-like-data 0 0 3

heegyu / wildguardmix-direct-1001 0 0 1

gabrielmbmb / ifeval-trl 0 0 1

orbench-llm / or-bench 0 0 1

HuggingFaceH4 / ifeval-like-data 0 0 3

taozi555 / Llama-Guard 0 0 1

ilsp / ifeval_greek 0 0 1

davisrbr / JBB-behaviors-embeddings-all-mpnet-base-v2 0 0 2

gabrielmbmb / ifeval-like-data-report 0 0 2

gabrielmbmb / ifeval-like-data-decon 0 0 1

le-leadboard / IFEval-fr 0 0 4

facebook / Multi-IF 0 0 4

iknow-lab / wildguardmix-train-ko-11k-trl 0 0 1

olinguyen / uplimit-finetuningllm-week1 0 0 1

david9dragon9 / cvalues-chinese 0 0 1

iknow-lab / wildguardmix-train-ko 0 0 1

iknow-lab / wildguardmix-train-ko-trl 0 0 1

dleemiller / toxic-pairs 0 0 1

abbasfar / backdoor_attack_evaluation_dataset 0 0 1

skirdey-inflection / paraphrased-salad-bench 0 0 1

mizinovmv / ru_ifeval-like-data 0 0 1

coderchen01 / HarmfulGeneration-HarmBench 0 0 1

allenai / tulu-3-IF-augmented-on-policy-8b 0 0 1

davisrbr / jailbreakbench-goal-embeddings-augmented 0 0 1

davisrbr / jailbreakbench-goal-embeddings-artifacts 0 0 2

makcedward / openai-moderation 0 0 1

bogdanminko / wildguardmix-cleaned 0 0 2

thunder-research-group / SNU_Ko-IFEval gated 0 1

aisingapore / Instruction-Following-IFEval gated 0 2

InazakiSakuya / llama-guard-harmful 0 0 2

projecte-aina / IFEval_ca 0 0 7

BSC-LT / IFEval_es 0 0 4

bench-llms / or-bench-toxic-all 0 0 2

david9dragon9 / xs_rl_data 0 0 1

gaydmi / IFEval_tt gated 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI 0 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI-hard-no 0 0 1

jkazdan / gemma-2-9b-it-cipher-gen3-harmless-5000-harmful-500-HeX-PHI-AMD 0 0 1

SicariusSicariiStuff / Magnum_ShareGPT 0 0 1

NoorNizar / HarmBench-Test-Standard 0 0 2

AlignmentResearch / XSTest 0 0 1

LLaMAX / BenchMAX_Rule-based 0 0 1

Ayush-Singh / IFEval_self_llama 0 0 1

Paul / XSTest 0 0 1

YinkaiW / harmbench-dataset 0 0 2

xchen16 / wjb_responses 0 0 1

ata990 / HarmBench_Base64 0 0 2

McGill-NLP / AdvBench-IR 0 0 2

kaleinaNyan / instruction-following-eval 0 0 1

hbseong / HarmAug_generated_dataset 0 0 2

usisoftware-org / JailbreakBench 0 0 2

AlignmentResearch / HarmBench 0 0 1

rasdani / smol-RLVR-IFeval 0 0 1

CocoRoF / IFEval-distill 0 0 1

llamastack / IfEval 0 0 1

inception42 / Arabic-IFEval 0 0 1

Delta-Vector / Tauri-IFeval-Dans-Tulu-KTO 0 0 1

tech4humans / JBB-Behaviors-pt 0 0 1

Jianshu001 / ifeval_dpsk 0 0 1

BrianGu / harmbench_behaviors 0 0 1

INSAIT-Institute / IFEval-bg 0 0 1

pxyyy / ifeval-like-data 0 0 1

pxyyy / Llama-3.1-8B-autoif-20k-1-1e-5-ifeval-correct-200 0 0 1

pxyyy / autoif-20k-removed-top1500-by-mp-for-ifeval-correct-200 0 0 1

agentlans / drill 0 0 1

pxyyy / autoif-20k-removed-top1500-by-mp-1e-4-for-ifeval-correct-200 0 0 1

allganize / IFEval-Ko 0 0 2

xsample / tulu-3-mig-50k 0 0 2

pxyyy / autoif-20k-removed-top1500-by-mp-1e-3-for-ifeval-correct-200 0 0 1

huihui-ai / harmbench_behaviors 0 0 1

pxyyy / autoif-20k-removed-top1000-by-mp-3e-4-for-ifeval-correct-200 0 0 1

ChrisDing1105 / MMIF-23k 0 0 1

xsample / tulu-3-ifd-50k 0 0 1

pxyyy / autoif-20k-removed-top7500-by-mp-1e-3-for-ifeval-correct-200 0 0 1

ai2-adapt-dev / tulu-3-thinker-classified-no_math-ifeval-ref-based-45k 0 0 1

AlignmentResearch / WildGuardTest 0 0 1

ai2-adapt-dev / tulu_3_rewritten_100k-v2-ifeval 0 0 1

AmazonScience / FalseReject 0 0 3

Beijing-AISI / panda-bench 0 0 2

ai2-adapt-dev / tulu_3_rewritten_100k-v3-ifeval-no_math 0 0 1

faezeb / tulu_3_rewritten_100k-v3-ifeval-general-no-ref 0 0 1

tungluuai / wildguard-multiple-questions gated 0 1

yixiaoh / OVERT 0 0 1

rasdani / ifeval-genesys 0 0 1

jdineen / or-bench 0 0 1

ai2-adapt-dev / tulu_3_rewritten_100k-v2-ifeval-replaced-math 0 0 1

BrachioLab / jailbreak_harmbench50 0 0 2

faezeb / RLVR-IFeval 0 0 1

iagoalves / wildguardmix_Qwen3-8B 0 0 1

iagoalves / wildguardmix_eval 0 0 1

chuhac / R2D-R1 0 0 1

zjunlp / ChineseHarm-bench 0 0 3

allenai / IF_multi_constraints_upto5 0 0 2

MatanBT / gcg-evaluated-data 0 0 2

iagoalves / wildguardmix_train_puro 0 0 1

JW17 / WildguardMix-non-ad-45k-Q25-1.5B-SFT 0 0 1

VGraf / ifeval_mt 0 0 1

japan-ai-official / jifeval 0 0 1

xunguangwang / JailbreakGuardrailBenchmark 0 0 2

BRlkl / WildGuardTest-pt 0 0 1

BRlkl / XSTest-pt 0 0 1

Sakonii / task-over-refusal-dataset 0 0 1

Roblox / RobloxGuard-Eval 0 0 2

Thanmay / IFEval 0 0 1

NoahShen / xstest-llama3.1-8b-inst-completions 0 0 1

alexanderstern / harmbench 0 0 1

dnouv / prompt_guardrail_eval 0 0 1

Thanmay / IFEval-verified 0 0 1

ilsp / Jailbreak-StrongReject-el 0 0 1

sa7270 / RLVR-IFEVAL-Filtered 0 0 1

purpcode / malevent-jailbreak-eval-oss 0 0 1

tartuNLP / ifeval_et 0 0 1

andyrdt / gpt-oss-20b-rollouts 0 0 2

oneonlee / wildguardmix-train 0 0 1

nvidia / Nemotron-Safety-Guard-Dataset-v3 0 0 3

DhruvTre / h2-harmbench-twins 0 0 1

jackzhang / JBDistill-Bench 0 0 1

Multilingual-Multimodal-NLP / IfEvalCode-testset 0 0 1

ibm-research / Split-IFEval 0 0 1

BRlkl / wildguardmix-pt 0 0 1

sbintuitions / WildGuardTestJP 0 0 2

m-a-p / Inverse_IFEval 0 0 2

DhruvTre / jailbreakbench-paraphrase-2025-08 0 0 2

evats / StrongReject_el 0 0 1

walledai / RoXSTEST gated 0 1

Oluwadara / JBB-Behaviors-harmful-nanogcg 0 0 1

Oluwadara / JBB-Behaviors 0 0 1

NoahShen / id-0011-xstest 0 0 1

manan-u / ifeval-translated 0 0 1

GeneralAnalysis / GA_Jailbreak_Benchmark gated 0 1

friendshipkim / IF-Datasets-Tulu-IFEval 0 0 1

NoahShen / id-0001-baseline-beear-harmbench 0 0 1

psyonp / SocialHarmBench 0 0 2

NikolaiSkripko / IFEval-FC 0 0 2

Thanmay / IFEval-sangraha-1 gated 0 1

Thanmay / IFEval-sangraha-2 gated 0 1

mikasenghaas / allenai-ifeval_gpt-5-mini_n5_r3 0 0 1

Jazhyc / wildguardmix-annotations 0 0 1

jaxon3062 / IFEval-gemma3-chat 0 0 1

boczkakaroly / trilingual-cultural-bias-redteaming-benchmark 0 0 1

shiv96 / HarmBench-standard-llama_3.2-1B-it-kpca-steered 0 0 1

sungyub / ifbench-verl 0 0 1

Lakera / b3-agent-security-benchmark-weak 0 0 2

sandbagging-games / evaluation_logs 0 0 1

marvintong / legal-llm-benchmark 0 0 1

centrepourlasecuriteia / content-moderation-input-dataset gated 0 2

puwaer / do_not_answer_en_response 0 0 1

NousResearch / eval-Hermes-4.3-36B-centralized 0 0 1

nvidia / Nemotron-Content-Safety-Reasoning-Dataset 0 0 1

AuroraQuantum / llama-guard-safety-eval 0 0 1

tartuNLP / ifeval_en 0 0 1

jbrin / or-bench-balanced 0 0 1

Neelectric / wildguardmix_Llama-3.1-8B-Instruct_4096toks gated 0 1

FlameF0X / Safety_Alignment_Benchmark gated 0 1

richardyoung / tempest-replication 0 0 1

denis1699 / jailbreak-benchmark-dataset-benign-advbench 0 0 1

denis1699 / jailbreak-benchmark-dataset-benign-gsm8k 0 0 1

sentinelseed / sentinel-evaluations 0 0 1

HarmlessSR07 / OSI-Bench 0 0 1

RebeccaYU920 / ifeval-pp 0 0 1

aiymka / JailbreakBench 0 0 1

Rendra86318 / IFEval 0 0 1

zhqwqwq / SFT_eval 0 0 1

agentlans / refusal-classifier-data 0 0 1

Aurther-Nadeem / introspect-ai-benchmark 0 0 1

pant0m / harmbench_classifier_train 0 0 1

CatBarks / harmbench_humanjailbreaks_llama2_strictjudge_success_only 0 0 1

yilingwang / OR-bench-classify 0 0 2

joaocarloscruz / chinese-refusals-benchmark 0 0 1

YuminChoi / ThinkSafe-8B-n4-filtered-LlamaGuard 0 0 1

rana-shahroz / ifeval-cot-llama3.3-70B 0 0 1

tonychenxyz / ifeval 0 0 1

zyL2003 / JBB-Behaviors 0 0 1

lamhots / BatakJailbreakBench 0 0 1

haralab-uec / steering-bench-ja 0 0 2

jamesbowen / JBB-Behaviors 0 0 1

Seanie-lee / ThinkSafe-0.6B-WildGuard 0 0 1

icml2026-7516 / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 1

ramendik / kimify-ifeval-like 0 0 1

icml2026-7516 / ThinkSafe-R1-Distill-7B-WildGuard 0 0 1

Neelectric / Replay_0.05.MoT_science.wildguardmix.Llama3_4096toks 0 0 1

Neelectric / wildguardmix_reasoning_Llama-3.1-8B-Instruct_4096toks 0 0 1

Seanie-lee / ThinkSafe-Qwen3-0.6B-WildGuard 0 0 1

Seanie-lee / ThinkSafe-Qwen3-1.7B-WildGuard 0 0 1

Seanie-lee / ThinkSafe-Qwen3-4B-WildGuard 0 0 2

Seanie-lee / ThinkSafe-Qwen3-8B-WildGuard 0 0 1

rahul-singhal / IFEval-Hard 0 0 1

Abel-24 / HarmMetric_Eval 0 0 1

Neelectric / wildguardmix_reasoning_Llama-3.1-8B-Instruct_4096toks_refusals_only 0 0 1

Neelectric / Replay_0.1.MoT_science.wildguardmix.Llama3_4096toks 0 0 1

acartag7 / gap-benchmark 0 0 1

jmk9494 / moderation-bias-benchmark 0 0 1

danish-foundation-models / multi-ifeval 0 0 1

mohdelgaar / ifeval_rlvr 0 0 1

eve6352 / RobloxGuard-Eval 0 0 1

ai4bharat / IndicIFEval 0 0 2

IanSteenstra / AI-Psychotherapy-Eval 0 0 1

haritzpuerto / ifeval-lrm 0 0 1

eazzy / wildguard_for_distil 0 0 1

LLMSafety / HarmBench 0 0 1

Faruna01 / lsr-benchmark 0 0 1

Hhhhhl / JBB-Behaviors 0 0 1

npow / memshield-bench 0 0 1

responsible-ai-labs / indian-responsible-ai-benchmark 0 0 1

jbross-ibm-research / M-IFEval 0 0 1

HiTZ / ifeval_eu 0 0 1

shuyilin / JO-Bench 0 0 1

FearedFusionX / RobloxGuard-Eval 0 0 1

minyichen / IF-R1 gated 0 1

timtkddn / Guardrail-eval-xstest-beavertail-30k-en-ko gated 0 1

HAC-Studios-Org / prompted-hearts-clinical-eval 0 0 1

ssu-csec / Role-SafetyBench 0 0 1

safety-aya / harmbench-portuguese 0 0 1

Tokyomonster / JBB-Behaviors 0 0 1

silmarilsecurity / firewall-eval-v1 gated 0 1

adityabagayatkar / JBB-Behaviors 0 0 1

Micdejc / cymultenset 0 0 1

intherejeet / agentic-rag-redteam-bench gated 0 1

davidquicast / sorry-bench-202503-multilingual-backup 0 0 1

HJH2CMD / lrm-safety-eval 0 0 1

boolishs / xstest-v2-copy 0 0 1

kairawal / MultiLingual-SorryBench 0 0 1

dmody1 / wildguardmix-refusal-generations 0 0 1

agentic-moral-alignment / gt-harmbench-eval 0 0 1

treadon / abliteration-eval 0 0 1

prithivMLmods / harm_bench 0 0 1

PowCal / HarmBench100 0 0 1

jang1563 / bio-overrefusal-v0.1 0 0 1

imadreamerboy / constitutional-safety-classifier-data 0 0 1

zzyppp / ifeval 0 0 2

stablegradients / IF-Datasets-Tulu-IFEval 0 0 1

CapyStudios / capydataset 0 0 1

chloeclerc17 / English_French_safety_code-mixing_dataset 0 0 1

choiqs / IF-Datasets-Tulu-IFEval 0 0 1

Jinesis / gt-harmbench 0 0 1

CaiZhiTech / Evaluation-Dataset-of-AI-Agent-Security-Guardrails 0 0 1

SamsungSDS-Research / Policy-on-the-Fly-Benchmark 0 0 1

treadon / disinhibition-eval 0 0 1

UCSC-VLAA / CIK-Bench 0 0 1

Jocana / herald-logits 0 0 1

memo-ozdincer / jepadata 0 0 1

samuelsimko / rr-circuit-breakers-attack-completions gated 0 1

bridge-benchmark-2026 / BRIDGE 0 0 1

neurips2026-10977 / ThinkSafe-Qwen3-4B-WildGuard 0 0 1

abliterationai / synthetic-data-provenance-eval 0 0 1

Panuthep / wildguardmix_2k 0 0 1

labp2d / UnifiedBench 0 0 1

jkminder / model-raising-pb-300k-3c-sft 0 0 1

jkminder / model-raising-pb-100k-3c-mt-sft 0 0 1

VityaVitalich / or-bench-1k 0 0 1

orq / redteam-framework-benchmark 0 0 1

jacobmorrison / cse-579-ifeval-rl 0 0 1

cs-552-2026-vibe-trainers / mcq_safety 0 0 1

Thanmay / IFEval-pipe 0 0 1

jkminder / or-bench-1k-overrefusal 0 0 1

massaindustries / dataset-A-routing 0 0 1

de-Rodrigo / ifeval-format-clean 0 0 1

anonymous-review-anonymous / HarmMetric_Eval 0 0 1

de-Rodrigo / ifeval-decontaminate 0 0 1

regolo / brick2-dataset-a-eval 0 0 2

m-zest / trojanspec-bench 0 0 1

khaledyusuf44 / somalibench-v0 0 0 1

szzh12 / JBB-Behaviors 0 0 1

pawel04 / ifeval-pl 0 0 1

pawel04 / ifeval-pl-200 0 0 1

Albertmade / nrt-bench gated 0 1

zzha6204 / Health-ORSC-Bench 0 0 1

zdxnlp / SaladBench 0 0 1

EliasHossain / ptc-benchmark 0 0 1

balabalab / JBB-Behaviors 0 0 1

hirundo-io / or-bench 0 0 1

anonymous-emnlp-2026-ejt / EJT_Benchmark gated 0 1

kasadin / bif-qg-safe-top120-eval-inference 0 0 1

inclusionAI / A3S-Bench 0 0 1

kasadin / alignment-4type 0 0 1

Jaemo3123 / JBB-Behaviors 0 0 1

sci4ai / Qwen3.5-27B-Abliterated-HarmBench-Refusals 0 0 1

hivetrace / harm_bench_completions gated 0 1

cs-552-2026-databand / Safety_Dataset_MCQ 0 0 1

cs-552-2026-qwenlifegivesyoulemons / safety_aegis_mcqa 0 0 1

moon12321 / WildGuardTest 0 0 1

PhilippBlue / mcq_safety_boxed 0 0 1

luna-pragma-2026 / Decontam-Bench 0 0 1

stindardlogic / refusal-overrefusal-50k 0 0 1

solanaclawd / solana-clawd-eval 0 0 1

FinancialSupport / IFEval-max-tokens 0 0 1

kurakurai / IFEval-FR 0 0 1

kevin-giskard / XSTest 0 0 1

lego17440 / guardrail-input-slm-data 0 0 1

UltraMar1ne / residual-streams-20260618 gated 0 1

giskardai / do-not-answer-scenarios 0 0 1

uninhibited-scholar / defensive-refusal-bench-zh 0 0 1

yavuz-ai / seas-coevolve 0 0 1

jash-ai / agentic-redteam-benchmark 0 0 1

PowerBench / PowerBench-Multilingual 0 0 1

anna-sarvam / indic-safety-eval 0 0 1

HeliumTrades / helium-model-worldview-benchmark 0 0 9

hirundo-io / bloom-over-refusal-free-text 0 0 1

ole-tail / agentic-execution-guardrail-eval 0 0 1

giskardai / harmbench-scenarios 0 0 1

SabrinaSadiekh / responses-and-asr-labels-small-models 0 0 1

marcodsn / crucible 0 0 1

ariffazil / a2b-eval-results 0 0 1

mahdieh-sjp / SafeRole 0 0 1

mirobody / MedHarm-Bench 0 0 1

nvidia / Nemotron-Content-VISafe-v1 0 0 2

lefft / needleif-bench 0 0 1

Tomk187 / LongHarmBench gated 0 1

SEBK4C / gemma4-serving-bench-data 0 0 1

astroware / halo-guard-bench 0 0 1

youbin2014 / PromptSecurity-Eval 0 0 2

devicemark / results 0 0 1

lamduong / human-ai-impact-bench-scenarios 0 0 1

ThePioneer / liberator-dataset 0 0 1

zmsy / compositional-safety-folds 0 0 1

Goutam112 / prompt-safety-classifier-100k 0 0 1

jerogo / or-bench 0 0 1

Magurofg / ifeval-pt 0 0 1

Cyber-security-final-project / Evaluation_of_OpenSource_Models_for_PDF_Injection_Recognition 0 0 1

fevziegeyurtsevenler / PolyGuardBench 0 0 1

ground-truth / bfsi-bench gated 0 1

digitable-lol / digit-eval-tasks 0 0 1

rustem17 / em-code-subliminal-transfer-evals 0 0 1

BushNate / llm-refusal-evaluation 0 0 1

agentdock / steerbench-work-2026-05 0 0 1

saintsauce / uniar-steering-eval 0 0 1

Qwovadis / red-team-appsec-benchmark 0 0 1

shrugging-shoulders / ifeval_multilang 0 0 1

hkab / vi-ifeval 0 0 1

MatanBT / JBRL-robustness-eval gated 0 1

lilyzhng / c-guard 0 0 1

heisenberg-88 / indirectrag-bench 0 0 1

kosiasuzu / tulu-postraining-data 0 0 1

Ishowbackup / JBB-Behaviors 0 0 1

MasihM / eyes-wide-shut-safety-benchmark 0 0 1

zakarth / ifeval-mini-120 0 0 1

groovinandmovin / JBB-Behaviors 0 0 1

jensjepsen / danish-if-grpo-combined-v3 0 0 1

usezn / zn-prompt-injection-bench 0 0 1

Siye01 / LLM-Fusion-Test 0 0 1

rlundqvist / ifeval-obf-rl-run 0 0 1

rlundqvist / ifeval-obf-rl-run-leak 0 0 1

adityanawal23 / HarmURLBench gated 0 1

HeraFox-ai / Mental-Health-Safety-Eval 0 0 4

hamiejuice / qwen3.8-27b-uncensored-dflash2-m4-pro-benchmark 0 0 1

frangelbarrera / ling-3-flash-evaluation 0 0 1

ruchit11111 / coding-agent-security-benchmark 0 0 1

shrugging-shoulders / inverse_ifeval_multilang 0 0 1

eda-bench-neurips-2026 / eda-bench-public-provenance 0 0 1

nanimani / local-llm-benchmark 0 0 1

ecairol / simpleqa-eval-qwen3.8-27b-obliterated 0 0 2

abeeranajam31 / CLSG-Benchmark 0 0 1

leexiaoyi-lab / HDHC-Benchmark 0 0 1

phionyx / airep-embedded-evaluation-profile 0 0 2

Nailyk14 / prompt-safety-multilingual 0 0 1

noahrossi / heretic-completions 0 0 1

CounterSteer / overrefusal-artifacts 0 0 1

shahoismael / prompt-quality-vs-response-compliance 0 0 1

Sangsang / ThinkSafe-Qwen3-8B-Activation-data 0 0 1

LeTG / psyop-bench 0 0 1

quickium / awall-eval-benchmark 0 0 1

mmse-bench-anon / msb 0 0 1

aakash-agarwal / Indic-Jailbreak-Bench gated 0 1

hanxunh / VEX-Bench 0 0 1

vhboyr / lexical-fear 0 0 1

Center-Of-Advanced-Software-Technologies / IFEval-Multi-IF-hy 0 0 1

ngila-ai / ngila-do-not-answer-v0.2 0 0 1

ValeEverheart / vale-governance-eval 0 0 1

Hastika06 / dualr-benchmark 0 just added 1

alfaplus / backdoor-detect-bench-v1 0 just added 1

submarat / mats11-mix4-readout2x 0 just added 1

RayNene / qwen3-multilingual-safety-eval 0 just added 1

← previous stage Heal All datasets next stage → Fine-tune

Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog