# reWordBench: Benchmarking and Improving the Robustness of Reward Models with Transformed Inputs

arXiv: https://arxiv.org/abs/2503.11751 · HTML: https://mlsanity.com/paper/2503.11751/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- Placed in Reward modeling and RLHF, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 63 of its 88 references that are in the map.
- First posted on arXiv 14 Mar 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 30 of 63)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Theoretically Principled Trade-off between Robustness and Accuracy](https://mlsanity.com/paper/1901.08573/) (2019)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity](https://mlsanity.com/paper/2104.08786/) (2021)
- [Scaling Laws for Reward Model Overoptimization](https://mlsanity.com/paper/2210.10760/) (2022)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Practical Black-Box Attacks against Machine Learning](https://mlsanity.com/paper/1602.02697/) (2016)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment](https://mlsanity.com/paper/2304.06767/) (2023)
- [Self-Rewarding Language Models](https://mlsanity.com/paper/2401.10020/) (2024)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment](https://mlsanity.com/paper/1907.11932/) (2019)
- [JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models](https://mlsanity.com/paper/2404.01318/) (2024)
- [Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study](https://mlsanity.com/paper/2305.13860/) (2023)
- [Learning the Difference that Makes a Difference with Counterfactually-Augmented Data](https://mlsanity.com/paper/1909.12434/) (2019)
- [How Far Can Camels Go? Exploring the State of Instruction Tuning on Open Resources](https://mlsanity.com/paper/2306.04751/) (2023)
- [UltraFeedback: Boosting Language Models with Scaled AI Feedback](https://mlsanity.com/paper/2310.01377/) (2023)
- [Fine-Grained Human Feedback Gives Better Rewards for Language Model Training](https://mlsanity.com/paper/2306.01693/) (2023)
- [Learning to Compress Prompts with Gist Tokens](https://mlsanity.com/paper/2304.08467/) (2023)
- [Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models](https://mlsanity.com/paper/2312.06585/) (2023)
- [Synthetic and Natural Noise Both Break Neural Machine Translation](https://mlsanity.com/paper/1711.02173/) (2017)
- [Evaluating Large Language Models at Evaluating Instruction Following](https://mlsanity.com/paper/2310.07641/) (2023)

## Built on by (most cited first, 25 of 25)

- [RewardBench 2: Advancing Reward Model Evaluation](https://mlsanity.com/paper/2506.01937/) (2025)
- [HelpSteer3-Preference: Open Human-Annotated Preference Data across Diverse Tasks and Languages](https://mlsanity.com/paper/2505.11475/) (2025)
- [Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2504.12501/) (2025)
- [Brittlebench: Quantifying LLM robustness via prompt sensitivity](https://mlsanity.com/paper/2603.13285/) (2026)
- [Robust Reward Modeling via Causal Rubrics](https://mlsanity.com/paper/2506.16507/) (2025)
- [Adversarial Training of Reward Models](https://mlsanity.com/paper/2504.06141/) (2025)
- [Incentivizing Agentic Reasoning in LLM Judges via Tool-Integrated Reinforcement Learning](https://mlsanity.com/paper/2510.23038/) (2025)
- [Enhancing Large Language Model Reasoning with Reward Models: An Analytical Survey](https://mlsanity.com/paper/2510.01925/) (2025)
- [Reward Auditor: Inference on Reward Modeling Suitability in Real-World Perturbed Scenarios](https://mlsanity.com/paper/2512.00920/) (2025)
- [Reward Modeling for Reinforcement Learning-Based LLM Reasoning: Design, Challenges, and Evaluation](https://mlsanity.com/paper/2602.09305/) (2026)
- [Activation Reward Models for Few-Shot Model Alignment](https://mlsanity.com/paper/2507.01368/) (2025)
- [Why is Your Language Model a Poor Implicit Reward Model?](https://mlsanity.com/paper/2507.07981/) (2025)
- [VideoRewardBench: Comprehensive Evaluation of Multimodal Reward Models for Video Understanding](https://mlsanity.com/paper/2509.00484/) (2025)
- [Libra: Assessing and Improving Reward Model by Learning to Think](https://mlsanity.com/paper/2507.21645/) (2025)
- [RoleRMBench & RoleRM: Towards Reward Modeling for Profile-Based Role Play in Dialogue Systems](https://mlsanity.com/paper/2512.10575/) (2025)
- [C2: Scalable Rubric-Augmented Reward Modeling from Binary Preferences](https://mlsanity.com/paper/2604.13618/) (2026)
- [Reward Model Perspectives: Whose Opinions Do Reward Models Reward?](https://mlsanity.com/paper/2510.06391/) (2025)
- [Fast-Slow Thinking RM: Efficient Integration of Scalar and Generative Reward Models](https://mlsanity.com/paper/2603.20212/) (2026)
- [What are the Right Symmetries for Formal Theorem Proving?](https://mlsanity.com/paper/2605.22257/) (2026)
- [Unifying Adversarial Robustness and Training Across Text Scoring Models](https://mlsanity.com/paper/2602.00857/) (2026)
- [On the Role of Citations in Preference Data](https://mlsanity.com/paper/2608.21376/) (2026)
- [Instruction Quality Matters: Refining Instructions for Effective Preference Learning](https://mlsanity.com/paper/2608.26779/) (2026)
- [MAWILE: Multi-Axis Workbench for Inspecting LLM Evaluators](https://mlsanity.com/paper/2609.22599/) (2026)
- [RMB: Reward Model Boosting Mitigates Reward Hacking](https://mlsanity.com/paper/2609.33221/) (2026)
- [Do MLLM Judges Judge the Edit? Auditing Bias in Image Editing Evaluation with Verified Quality Preservation](https://mlsanity.com/paper/2610.01670/) (2026)

## Abstract (arXiv)

Reward models have become a staple in modern NLP, serving as not only a scalable text evaluator, but also an indispensable component in many alignment recipes and inference-time algorithms. However, while recent reward models increase performance on standard benchmarks, this may partly be due to overfitting effects, which would confound an understanding of their true capability. In this work, we scrutinize the robustness of reward models and the extent of such overfitting. We build **reWordBench**, which systematically transforms reward model inputs in meaning- or ranking-preserving ways. We show that state-of-the-art reward models suffer from substantial performance degradation even with minor input transformations, sometimes dropping to significantly below-random accuracy, suggesting brittleness. To improve reward model robustness, 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. For example, our robust reward model reduces such degradation by roughly half for the Chat Hard subset in RewardBench. Furthermore, when used in alignment, our robust reward models demonstrate better utility and lead to higher-quality outputs, winning in up to 59% of instances against a standardly trained RM.