LRCU: DLM Weight Unlearning
Collection
LRCU weight-unlearning checkpoints for masked diffusion LMs (LLaDA-8B, Dream-7B) on WMDP-bio/cyber and RWKU, plus a TOFU SFT target. • 10 items • Updated
LRCU-unlearned llada-8b-tofu-sft (TOFU)
guanmingchiu/llada-8b-tofu-sft unlearned on TOFU with LRCU (Localized Recall-Capped Unlearning): a saturating per-token recall cap on a causally-localized block band (blocks 23-25) + bounded Min-SNR (low-$t$) trajectory weighting + CE retain anchor, with no reference model.
from transformers import AutoModel, AutoTokenizer
m = AutoModel.from_pretrained("guanmingchiu/lrcu-llada-8b-tofu", trust_remote_code=True)
t = AutoTokenizer.from_pretrained("guanmingchiu/lrcu-llada-8b-tofu", trust_remote_code=True)
- Downloads last month
- 7
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for guanmingchiu/lrcu-llada-8b-tofu
Base model
guanmingchiu/llada-8b-tofu-sft