paper-with-me

홈 › Papers

One-Shot Safety Alignment for Large Language Models via Optimal Dualization

2024-05-29 · Xinmeng Huang, Shuo Li, Edgar Dobriban, Osbert Bastani, Hamed Hassani, Dongsheng Ding

The growing safety concerns surrounding large language models raise an urgent need to align them with diverse human preferences to simultaneously enhance their helpfulness and safety. A promising approach is to enforce safety constraints through Reinforcement Learning from Human Feedback (RLHF). For such constrained RLHF, typical Lagrangian-based primal-dual policy optimization methods are computationally expensive and often unstable. This paper presents a perspective of dualization that reduces constrained alignment to an equivalent unconstrained alignment problem. We do so by pre-optimizing a smooth and convex dual function that has a closed form. This shortcut eliminates the need for cumbersome primal-dual policy iterations, greatly reducing the computational burden and improving training stability. Our strategy leads to two practical algorithms in model-based and preference-based settings (MoCAN and PeCAN, respectively). A broad range of experiments demonstrate the effectiveness and merits of our algorithms.

📄 PDF Abstract BibTeX arXiv:2405.19544

Code (1)

shuoli90/CAN 공식 구현 pytorch

Tasks

Safety Alignment

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

SciSafeEval: A Comprehensive Benchmark for Safety Alignment of Large Language Models in Scientific Tasks

2024-10-02 · TianHao Li, Jingyu Lu, Chuangxin Chu, Tianyu Zeng 외

Large language models (LLMs) have a transformative impact on a variety of scientific tasks across disciplines including biology, chemistry, medicine, and physics. However, ensuring the safety alignment of these models in…

Safety Alignment

Sycophancy as a Multilingual Alignment Failure: How Safety Degrades Across Languages, Topics, and Models

2026-06-07 · Arya Shah, Himanshu Beniwal, Mayank Singh, Chaklam Silpasuwanchai arxiv

Safety-aligned large language models often exhibit sycophancy, which is the tendency to affirm users' opinions regardless of factual accuracy. Although well-studied in English, its manifestation in other languages remain…

Stepwise Alignment for Constrained Language Model Policy Optimization

2024-04-17 · Akifumi Wachi, Thien Q. Tran, Rei Sato, Takumi Tanabe 외

Safety and trustworthiness are indispensable requirements for real-world applications of AI systems using large language models (LLMs). This paper formulates human value alignment as an optimization problem of the langua…

Computational EfficiencyLanguage ModelingLanguage Modellingmodel

GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher

2023-08-12 · Youliang Yuan, Wenxiang Jiao, Wenxuan Wang, Jen-tse Huang 외

Safety lies at the core of the development of Large Language Models (LLMs). There is ample work on aligning LLMs with human ethics and preferences, including data filtering in pretraining, supervised fine-tuning, reinfor…

EthicsRed TeamingSafety Alignment

Decoupling Safety into Orthogonal Subspace: Cost-Efficient and Performance-Preserving Alignment for Large Language Models

2025-10-10 · Yutao Mou, Xiaoling Zhou, Yuxiao Luo, Shikun Zhang 외 arxiv

Safety alignment is essential for building trustworthy artificial intelligence, yet it remains challenging to enhance model safety without degrading general performance. Current approaches require computationally expensi…