Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.24154 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911710694080512 |
|---|---|
| author | Tan, Qitao Song, Xiaoying Akbari, Arman Akbari, Arash Wang, Yanzhi Zhai, Xiaoming Hong, Lingzi Xiang, Zhen Lu, Jin Yuan, Geng |
| author_facet | Tan, Qitao Song, Xiaoying Akbari, Arman Akbari, Arash Wang, Yanzhi Zhai, Xiaoming Hong, Lingzi Xiang, Zhen Lu, Jin Yuan, Geng |
| contents | Current safety alignment of foundation models largely follows a \emph{one-size-fits-all} paradigm, applying the same refusal policy across users and contexts. As a result, models may refuse requests that are unsafe for general users but legitimate for authorized professionals, limiting helpfulness in specialized professional settings. Existing approaches either require costly realignment or rely on inference-time steering that suffers from imprecise control and added latency. To this end, we propose \textsc{Palette}, a modular, controllable, and efficient framework that selectively relaxes refusal behavior on authorized target domains while preserving standard safety elsewhere. Our method identifies a refusal direction via multi-objective search and internalizes it into the model through lightweight adaptation. \textsc{Palette} further supports modular composition: it learns domain-specific safety controls independently and composes them through parameter merging, enabling on-demand multi-domain authorization without retraining. Experiments across four safety benchmarks, multiple model variants, and both LLMs and VLMs show that \textsc{Palette} delivers precise safety control without sacrificing general utility, offering a practical path toward foundation models that adapt to diverse professional needs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_24154 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs Tan, Qitao Song, Xiaoying Akbari, Arman Akbari, Arash Wang, Yanzhi Zhai, Xiaoming Hong, Lingzi Xiang, Zhen Lu, Jin Yuan, Geng Artificial Intelligence Software Engineering Current safety alignment of foundation models largely follows a \emph{one-size-fits-all} paradigm, applying the same refusal policy across users and contexts. As a result, models may refuse requests that are unsafe for general users but legitimate for authorized professionals, limiting helpfulness in specialized professional settings. Existing approaches either require costly realignment or rely on inference-time steering that suffers from imprecise control and added latency. To this end, we propose \textsc{Palette}, a modular, controllable, and efficient framework that selectively relaxes refusal behavior on authorized target domains while preserving standard safety elsewhere. Our method identifies a refusal direction via multi-objective search and internalizes it into the model through lightweight adaptation. \textsc{Palette} further supports modular composition: it learns domain-specific safety controls independently and composes them through parameter merging, enabling on-demand multi-domain authorization without retraining. Experiments across four safety benchmarks, multiple model variants, and both LLMs and VLMs show that \textsc{Palette} delivers precise safety control without sacrificing general utility, offering a practical path toward foundation models that adapt to diverse professional needs. |
| title | Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs |
| topic | Artificial Intelligence Software Engineering |
| url | https://arxiv.org/abs/2605.24154 |