---
license: mit
base_model: GSAI-ML/LLaDA-8B-Instruct
tags: [unlearning, diffusion-language-model, wmdp, safety]
---

# LRCU-unlearned LLaDA-8B (WMDP-bio)

`GSAI-ML/LLaDA-8B-Instruct` unlearned on WMDP-bio hazardous knowledge with LRCU
(Localized Recall-Capped Unlearning): a saturating per-token recall cap on a
causally-localized block band (blocks 5-7) + bounded Min-SNR (low-$t$) trajectory
weighting + CE retain anchor, with no reference model.

WMDP-bio accuracy drops to near chance while MMLU is largely preserved.

```python
from transformers import AutoModel, AutoTokenizer
m = AutoModel.from_pretrained("guanmingchiu/lrcu-llada-8b-wmdp-bio", trust_remote_code=True)
t = AutoTokenizer.from_pretrained("guanmingchiu/lrcu-llada-8b-wmdp-bio", trust_remote_code=True)
```