uiuc-kang-lab/PersistBD
Alignment dataset tracked in the /datasets sub-catalog.
Description
PersistBD Dataset
This dataset accompanies the paper "Understanding and Enhancing Backdoor Persistency in
LLM Agent Post-Training" (EMNLP 2026 Findings).
Code: github.com/uiuc-kang-lab/PersistBD.
⚠️ This data is poisoned — do not train on it unintentionally
A subset of these trajectories is backdoored: a trigger comment # |TEST MODE|
inserted into the agent's input elicits a malicious tool call that exfiltrates a repo's
.env file to an attacker-controlled… See the full description on the dataset page: https://huggingface.co/datasets/uiuc-kang-lab/PersistBD.
Tags
Used by abliterated models
No abliterated models in our catalog have declared this dataset in their YAML metadata yet. This can mean the dataset is used but not declared, is used outside abliteration workflows, or is new. The link will populate automatically as models are indexed.
Related datasets
Read further
Metadata
Download the dataset
huggingface-cli login and use one of the snippets below.
Copy any of these snippets into your notebook or terminal. All three fetch directly from Hugging Face using your own credentials.
from datasets import load_dataset
ds = load_dataset("uiuc-kang-lab/PersistBD")from huggingface_hub import snapshot_download
snapshot_download(repo_id="uiuc-kang-lab/PersistBD", repo_type="dataset")git clone https://huggingface.co/datasets/uiuc-kang-lab/PersistBD