RM-Bench: Benchmarking Reward Models of Language Models with Subtlety and Style
- URL: http://arxiv.org/abs/2410.16184v1
- Date: Mon, 21 Oct 2024 16:48:26 GMT
- Title: RM-Bench: Benchmarking Reward Models of Language Models with Subtlety and Style
- Authors: Yantao Liu, Zijun Yao, Rui Min, Yixin Cao, Lei Hou, Juanzi Li,
- Abstract summary: RM-Bench is a novel benchmark designed to evaluate reward models based on their sensitivity to subtle content differences and resistance to style biases.
We evaluate nearly 40 reward models on RM-Bench and find that even state-of-the-art models achieve an average performance of only 46.6%.
- Score: 37.97757796124621
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Reward models are critical in techniques like Reinforcement Learning from Human Feedback (RLHF) and Inference Scaling Laws, where they guide language model alignment and select optimal responses. Despite their importance, existing reward model benchmarks often evaluate models by asking them to distinguish between responses generated by models of varying power. However, this approach fails to assess reward models on subtle but critical content changes and variations in style, resulting in a low correlation with policy model performance. To this end, we introduce RM-Bench, a novel benchmark designed to evaluate reward models based on their sensitivity to subtle content differences and resistance to style biases. Extensive experiments demonstrate that RM-Bench strongly correlates with policy model performance, making it a reliable reference for selecting reward models to align language models effectively. We evaluate nearly 40 reward models on RM-Bench. Our results reveal that even state-of-the-art models achieve an average performance of only 46.6%, which falls short of random-level accuracy (50%) when faced with style bias interference. These findings highlight the significant room for improvement in current reward models. Related code and data are available at https://github.com/THU-KEG/RM-Bench.
Related papers
- Activation Reward Models for Few-Shot Model Alignment [77.37511364793515]
We introduce Activation Reward Models (Activation RMs)<n>Activation RMs leverage activation steering to construct well-aligned reward signals using minimal supervision and no additional model finetuning.<n>We demonstrate the effectiveness of Activation RMs in mitigating reward hacking behaviors, highlighting their utility for safety-critical applications.
arXiv Detail & Related papers (2025-07-02T05:10:29Z) - Reward Model Interpretability via Optimal and Pessimal Tokens [4.951383975460995]
Reward modeling has emerged as a crucial component in aligning large language models with human values.<n>We present a novel approach to reward model interpretability through exhaustive analysis of their responses across their entire vocabulary space.<n>We find that these models can encode concerning biases toward certain identity groups, which may emerge as unintended consequences of harmlessness training.
arXiv Detail & Related papers (2025-06-08T23:56:58Z) - Discriminative Policy Optimization for Token-Level Reward Models [55.98642069903191]
Process reward models (PRMs) provide more nuanced supervision compared to outcome reward models (ORMs)<n>Q-RM explicitly learns token-level Q-functions from preference data without relying on fine-grained annotations.<n>Reinforcement learning with Q-RM significantly enhances training efficiency, achieving convergence 12 times faster than ORM on GSM8K and 11 times faster than step-level PRM on MATH.
arXiv Detail & Related papers (2025-05-29T11:40:34Z) - Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization [15.729285736811383]
Reward models play a crucial role in reinforcement learning from human feedback.<n>Existing benchmarks for reward models show a weak correlation with the performance of optimized policies.
arXiv Detail & Related papers (2025-05-19T06:43:08Z) - RM-R1: Reward Modeling as Reasoning [81.50471199906738]
Reasoning Reward Models (ReasRMs) formulate reward modeling as a reasoning task.<n>We propose a reasoning-oriented training pipeline and train a family of ReasRMs, RM-R1.<n>Our models achieve state-of-the-art performance across three reward model benchmarks on average.
arXiv Detail & Related papers (2025-05-05T06:11:12Z) - reWordBench: Benchmarking and Improving the Robustness of Reward Models with Transformed Inputs [64.29893431743608]
We show that state-of-the-art reward models suffer from substantial performance degradation even with minor input transformations.
We propose to explicitly train them to assign similar scores to paraphrases, and find that this approach also improves robustness to other distinct kinds of transformations.
arXiv Detail & Related papers (2025-03-14T17:59:41Z) - CARMO: Dynamic Criteria Generation for Context-Aware Reward Modelling [27.86204841898399]
Reward modeling in large language models is susceptible to reward hacking.
We propose Context-Aware Reward Modeling (CARMO) to mitigate this problem.
We establish a new state-of-the-art performance in zero-shot settings for generative models, achieving a 2.1% improvement on Reward Bench.
arXiv Detail & Related papers (2024-10-28T21:18:49Z) - Evaluating Robustness of Reward Models for Mathematical Reasoning [14.97819343313859]
We introduce a new design for reliable evaluation of reward models, and to validate this, we construct RewardMATH.
We demonstrate that the scores on RewardMATH strongly correlate with the results of optimized policy and effectively estimate reward overoptimization.
arXiv Detail & Related papers (2024-10-02T16:39:58Z) - Quantile Regression for Distributional Reward Models in RLHF [1.8130068086063336]
We introduce Quantile Reward Models (QRMs), a novel approach to reward modeling that learns a distribution over rewards instead of a single scalar value.
Our method uses quantile regression to estimate a full, potentially multimodal distribution over preferences, providing a more powerful and nuanced representation of preferences.
Our experimental results show that QRM outperforms comparable traditional point-estimate models on RewardBench.
arXiv Detail & Related papers (2024-09-16T10:54:04Z) - Critique-out-Loud Reward Models [20.631830494414096]
We introduce Critique-out-Loud (CLoud) reward models.
CLoud reward models operate by first generating a natural language critique of the assistant's response.
We demonstrate the success of CLoud reward models for both Llama-3-8B and 70B base models.
arXiv Detail & Related papers (2024-08-21T17:24:15Z) - HAF-RM: A Hybrid Alignment Framework for Reward Model Training [51.59246299566669]
We propose a hybrid alignment framework HaF-RM for reward model training.
It offers a principled and effective approach to enhancing the performance and alignment of reward models.
arXiv Detail & Related papers (2024-07-04T23:26:56Z) - RewardBench: Evaluating Reward Models for Language Modeling [100.28366840977966]
We present RewardBench, a benchmark dataset and code-base for evaluation of reward models.
The dataset is a collection of prompt-chosen-rejected trios spanning chat, reasoning, and safety.
On the RewardBench leaderboard, we evaluate reward models trained with a variety of methods.
arXiv Detail & Related papers (2024-03-20T17:49:54Z) - Secrets of RLHF in Large Language Models Part II: Reward Modeling [134.97964938009588]
We introduce a series of novel methods to mitigate the influence of incorrect and ambiguous preferences in the dataset.
We also introduce contrastive learning to enhance the ability of reward models to distinguish between chosen and rejected responses.
arXiv Detail & Related papers (2024-01-11T17:56:59Z) - Investigating Ensemble Methods for Model Robustness Improvement of Text
Classifiers [66.36045164286854]
We analyze a set of existing bias features and demonstrate there is no single model that works best for all the cases.
By choosing an appropriate bias model, we can obtain a better robustness result than baselines with a more sophisticated model design.
arXiv Detail & Related papers (2022-10-28T17:52:10Z) - On the model-based stochastic value gradient for continuous
reinforcement learning [50.085645237597056]
We show that simple model-based agents can outperform state-of-the-art model-free agents in terms of both sample-efficiency and final reward.
Our findings suggest that model-based policy evaluation deserves closer attention.
arXiv Detail & Related papers (2020-08-28T17:58:29Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.