FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language Model
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908604634759168 |
|---|---|
| author | Hu, Jinwei Huang, Zhenglin Yin, Xiangyu Ruan, Wenjie Cheng, Guangliang Dong, Yi Huang, Xiaowei |
| author_facet | Hu, Jinwei Huang, Zhenglin Yin, Xiangyu Ruan, Wenjie Cheng, Guangliang Dong, Yi Huang, Xiaowei |
| contents | Large language models have been widely applied, but can inadvertently encode sensitive or harmful information, raising significant safety concerns. Machine unlearning has emerged to alleviate this concern; however, existing training-time unlearning approaches, relying on coarse-grained loss combinations, have limitations in precisely separating knowledge and balancing removal effectiveness with model utility. In contrast, we propose Fine-grained Activation manipuLation by Contrastive Orthogonal uNalignment (FALCON), a novel representation-guided unlearning approach that leverages information-theoretic guidance for efficient parameter selection, employs contrastive mechanisms to enhance representation separation, and projects conflict gradients onto orthogonal subspaces to resolve conflicts between forgetting and retention objectives. Extensive experiments demonstrate that FALCON achieves superior unlearning effectiveness while maintaining model utility, exhibiting robust resistance against knowledge recovery attempts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_01472 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language Model Hu, Jinwei Huang, Zhenglin Yin, Xiangyu Ruan, Wenjie Cheng, Guangliang Dong, Yi Huang, Xiaowei Computation and Language Artificial Intelligence Large language models have been widely applied, but can inadvertently encode sensitive or harmful information, raising significant safety concerns. Machine unlearning has emerged to alleviate this concern; however, existing training-time unlearning approaches, relying on coarse-grained loss combinations, have limitations in precisely separating knowledge and balancing removal effectiveness with model utility. In contrast, we propose Fine-grained Activation manipuLation by Contrastive Orthogonal uNalignment (FALCON), a novel representation-guided unlearning approach that leverages information-theoretic guidance for efficient parameter selection, employs contrastive mechanisms to enhance representation separation, and projects conflict gradients onto orthogonal subspaces to resolve conflicts between forgetting and retention objectives. Extensive experiments demonstrate that FALCON achieves superior unlearning effectiveness while maintaining model utility, exhibiting robust resistance against knowledge recovery attempts. |
| title | FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language Model |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2502.01472 |