paper-with-me

Papers

Representation-based Reward Modeling for Efficient Safety Alignment of Large Language Model

2025-03-13 · Qiyuan Deng, Xuefeng Bai, Kehai Chen, YaoWei Wang, Liqiang Nie, Min Zhang

Reinforcement Learning (RL) algorithms for safety alignment of Large Language Models (LLMs), such as Direct Preference Optimization (DPO), encounter the challenge of distribution shift. Current approaches typically address this issue through online sampling from the target policy, which requires significant computational resources. In this paper, we hypothesize that during off-policy training, while the ranking order of output generated by policy changes, their overall distribution remains relatively stable. This stability allows the transformation of the sampling process from the target policy into a re-ranking of preference data. Building on this hypothesis, We propose a new framework that leverages the model's intrinsic safety judgment capability to extract reward signals, which are then used to calculate label confidence for preferences reordering. Extensive experimental results and theoretical analysis demonstrate that the proposed method effectively addresses the distribution shift issue, remarkably enhancing the safety performance while reducing about 300x computational overheads.

📄 PDF Abstract BibTeX arXiv:2503.10093

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage ModellingLarge Language ModelReinforcement Learning (RL)Re-RankingSafety Alignment

Similar Papers 제목 키워드 기반

Configurable Reward Model for Balanced Safety Alignment

2026-05-28 · Zhengping Jiang, Mehran Khodabandeh, Akash Bharadwaj, Manik Bhandari 외 arxiv

Aligning large language models (LLMs) to heterogeneous and rapidly evolving safety requirements remains a critical challenge. Existing instruction-tuned LLMs and standalone safety classifiers often fail to generalize to …

Data Augmentation

Alignment and Safety of Diffusion Models via Reinforcement Learning and Reward Modeling: A Survey

2025-05-23 · Preeti Lamba, Kiran Ravish, Ankita Kushwaha, Pawan Kumar

Diffusion models have emerged as leading generative models for images and other modalities, but aligning their outputs with human preferences and safety constraints remains a critical challenge. This thesis proposal inve…

Active LearningReinforcement Learning (RL)Safety Alignment

Learning from Reference Answers: Versatile Language Model Alignment without Binary Human Preference Data

2025-04-14 · Shuai Zhao, Linchao Zhu, Yi Yang

Large language models~(LLMs) are expected to be helpful, harmless, and honest. In various alignment scenarios, such as general human preference, safety, and confidence alignment, binary preference data collection and rew…

Language ModelingLanguage Modelling

Activation Reward Models for Few-Shot Model Alignment

2025-07-02 · Tianning Chai, Chancharik Mitra, Brandon Huang, Gautam Rajendrakumar Gare 외 arxiv

Aligning Large Language Models (LLMs) and Large Multimodal Models (LMMs) to human preferences is a central challenge in improving the quality of the models' generative outputs for real-world applications. A common approa…

Reinforcement Learning

Approximated Variational Bayesian Inverse Reinforcement Learning for Large Language Model Alignment

2024-11-14 · Yuang Cai, Yuyu Yuan, Jinsheng Shi, Qinhong Lin

The alignment of large language models (LLMs) is crucial for generating helpful and harmless content. Existing approaches leverage preference-based human feedback data to learn the reward function and align the LLM with …

BIRLImitation LearningLanguage ModelingLanguage Modelling+2