paper-with-me

Papers

RRM: Robust Reward Model Training Mitigates Reward Hacking

2024-09-20 · Tianqi Liu, Wei Xiong, Jie Ren, Lichang Chen, Junru Wu, Rishabh Joshi, Yang Gao, Jiaming Shen, Zhen Qin, Tianhe Yu, Daniel Sohn, Anastasiia Makarova, Jeremiah Liu, YuAn Liu, Bilal Piot, Abe Ittycheriah, Aviral Kumar, Mohammad Saleh

Reward models (RMs) play a pivotal role in aligning large language models (LLMs) with human preferences. However, traditional RM training, which relies on response pairs tied to specific prompts, struggles to disentangle prompt-driven preferences from prompt-independent artifacts, such as response length and format. In this work, we expose a fundamental limitation of current RM training methods, where RMs fail to effectively distinguish between contextual signals and irrelevant artifacts when determining preferences. To address this, we introduce a causal framework that learns preferences independent of these artifacts and propose a novel data augmentation technique designed to eliminate them. Extensive experiments show that our approach successfully filters out undesirable artifacts, yielding a more robust reward model (RRM). Our RRM improves the performance of a pairwise reward model trained on Gemma-2-9b-it, on RewardBench, increasing accuracy from 80.61% to 84.15%. Additionally, we train two DPO policies using both the RM and RRM, demonstrating that the RRM significantly enhances DPO-aligned policies, improving MT-Bench scores from 7.27 to 8.31 and length-controlled win-rates in AlpacaEval-2 from 33.46% to 52.49%.

📄 PDF Abstract BibTeX arXiv:2409.13156

Code (0)

등록된 구현이 없습니다.

Tasks

Data Augmentationmodel

Methods 이 논문이 사용한 방법론

DPO 설명 없음

Similar Papers 제목 키워드 기반

Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor Correction

2025-08-06 · Ruike Song, Zeen Song, Huijie Guo, Wenwen Qiang arxiv

External reasoning systems combine language models with process reward models (PRMs) to select high-quality reasoning paths for complex tasks such as mathematical problem solving. However, these systems are prone to rewa…

Causal Inference

The Horcrux: Mechanistically Interpretable Task Decomposition for Detecting and Mitigating Reward Hacking in Embodied AI Systems

2025-11-22 · Subramanyam Sahoo, Jared Junkin arxiv

Embodied AI agents exploit reward signal flaws through reward hacking, achieving high proxy scores while failing true objectives. We introduce Mechanistically Interpretable Task Decomposition (MITD), a hierarchical trans…

Directional Alignment Mitigates Reward Hacking in Reinforcement Learning for Language Models

2026-05-24 · Wenlong Deng, Jiaji Huang, Kaan Ozkara, Yushu Li 외 arxiv

Reward hacking arises when a model improves a proxy reward by exploiting shortcuts rather than solving the intended task. We study this failure mode through the geometry of reinforcement learning updates in language mode…

Reinforcement LearningMathematical Reasoning

Correlated Proxies: A New Definition and Improved Mitigation for Reward Hacking

2024-03-05 · Cassidy Laidlaw, Shivam Singhal, Anca Dragan

Because it is difficult to precisely specify complex objectives, reinforcement learning policies are often optimized using proxy reward functions that only approximate the true goal. However, optimizing proxy rewards fre…

reinforcement-learningReinforcement Learning

Cooper: Co-Optimizing Policy and Reward Models in Reinforcement Learning for Large Language Models

2025-08-07 · Haitao Hong, Yuchen Yan, Xingyu Wu, Guiyang Hou 외 arxiv

Large language models (LLMs) have demonstrated remarkable performance in reasoning tasks, where reinforcement learning (RL) serves as a key algorithm for enhancing their reasoning capabilities. Currently, there are two m…

Reinforcement Learning