Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance

• URL: http://arxiv.org/abs/2512.23461v1
• Date: Mon, 29 Dec 2025 13:39:41 GMT
• Title: Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance
• Authors: Zhuo Li, Pengyu Cheng, Zhechao Yu, Feifei Tong, Anningzhe Gao, Tsung-Hui Chang, Xiang Wan, Erchao Zhao, Xiaoxi Jiang, Guanjun Jiang,
• Abstract summary: Reward models (RMs) are essential in reinforcement learning from human feedback (RLHF) to align large language models with human values.<n>We introduce a novel information-theoretic debiasing method called textbfDebiasing via textbfInformation optimization for textbfRM (DIR)<n>With theoretical justification from information theory, DIR can handle more sophisticated types of biases with non-linear correlations.
• Score: 46.71732887299883
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reward models (RMs) are essential in reinforcement learning from human feedback (RLHF) to align large language models (LLMs) with human values. However, RM training data is commonly recognized as low-quality, containing inductive biases that can easily lead to overfitting and reward hacking. For example, more detailed and comprehensive responses are usually human-preferred but with more words, leading response length to become one of the inevitable inductive biases. A limited number of prior RM debiasing approaches either target a single specific type of bias or model the problem with only simple linear correlations, \textit{e.g.}, Pearson coefficients. To mitigate more complex and diverse inductive biases in reward modeling, we introduce a novel information-theoretic debiasing method called \textbf{D}ebiasing via \textbf{I}nformation optimization for \textbf{R}M (DIR). Inspired by the information bottleneck (IB), we maximize the mutual information (MI) between RM scores and human preference pairs, while minimizing the MI between RM outputs and biased attributes of preference inputs. With theoretical justification from information theory, DIR can handle more sophisticated types of biases with non-linear correlations, broadly extending the real-world application scenarios for RM debiasing methods. In experiments, we verify the effectiveness of DIR with three types of inductive biases: \textit{response length}, \textit{sycophancy}, and \textit{format}. We discover that DIR not only effectively mitigates target inductive biases but also enhances RLHF performance across diverse benchmarks, yielding better generalization abilities. The code and training recipes are available at https://github.com/Qwen-Applications/DIR.

Related papers

• Off-Policy Corrected Reward Modeling for Reinforcement Learning from Human Feedback [52.1410307583181]
  We useReinforcement Learning from Human Feedback to train language models (LMs) to follow complex human preferences.<n>As training progresses, the responses generated by the LM no longer resemble the responses seen by the reward model (RM)<n>We propose Off-Policy Corrected Reward Modeling to correct the RM using importance weighting, without requiring new labels or samples.
  arXiv  Detail & Related papers  (2025-07-21T11:19:04Z)
• Bias Fitting to Mitigate Length Bias of Reward Model in RLHF [81.44256822500257]
  Reinforcement Learning from Human Feedback relies on reward models to align large language models with human preferences.<n>We propose FiMi-RM, a framework that autonomously learns and corrects underlying bias patterns.<n> Experimental results demonstrate that FiMi-RM achieves a more balanced length-reward distribution.
  arXiv  Detail & Related papers  (2025-05-19T08:29:28Z)
• Disentangling Length Bias In Preference Learning Via Response-Conditioned Modeling [87.17041933863041]
  Reinforcement Learning from Human Feedback (RLHF) has achieved considerable success in aligning large language models (LLMs)<n>We introduce a $textbfR$esponse-$textbfc$onditioned $textbfB$radley-$textbfT$erry (Rc-BT) model that enhances the model's capability in length bias mitigating and length instruction following.<n>We also propose the Rc-RM and Rc-DPO algorithm to leverage the Rc-BT model for reward modeling and direct policy optimization
  arXiv  Detail & Related papers  (2025-02-02T14:50:25Z)
• Rethinking Relation Extraction: Beyond Shortcuts to Generalization with a Debiased Benchmark [53.876493664396506]
  Benchmarks are crucial for evaluating machine learning algorithm performance, facilitating comparison and identifying superior solutions.<n>This paper addresses the issue of entity bias in relation extraction tasks, where models tend to rely on entity mentions rather than context.<n>We propose a debiased relation extraction benchmark DREB that breaks the pseudo-correlation between entity mentions and relation types through entity replacement.<n>To establish a new baseline on DREB, we introduce MixDebias, a debiasing method combining data-level and model training-level techniques.
  arXiv  Detail & Related papers  (2025-01-02T17:01:06Z)
• Reward Modeling with Ordinal Feedback: Wisdom of the Crowd [9.034189257088762]
  Learning a reward model (RM) from human preferences has been an important component in aligning large language models. We propose a framework for learning RMs under ordinal feedback. We prove the statistical benefits of ordinal feedback in terms of reducing the Rademacher complexity.
  arXiv  Detail & Related papers  (2024-11-19T20:17:04Z)
• Post-hoc Reward Calibration: A Case Study on Length Bias [38.47276516266]
  Reward models (RMs) can develop biases by exploiting spurious correlations in their training data.<n>These biases can lead to incorrect output rankings, sub-optimal model evaluations, and the amplification of undesirable behaviours.<n>This paper addresses the challenge of correcting such biases without additional data and training.
  arXiv  Detail & Related papers  (2024-09-25T22:30:42Z)
• Reward Difference Optimization For Sample Reweighting In Offline RLHF [18.62836654699957]
  Current offline RLHF only captures the "ordinal relationship" between responses, overlooking the crucial aspect of how much one is preferred over the others. We propose a simple yet effective solution called Reward Difference Optimization, shorted as RDO. Experiments with 7B LLMs on the HH and TL;DR datasets substantiate the effectiveness of our method in both automatic metrics and human evaluation.
  arXiv  Detail & Related papers  (2024-08-18T07:04:16Z)
• Debiasing Multimodal Models via Causal Information Minimization [65.23982806840182]
  We study bias arising from confounders in a causal graph for multimodal data. Robust predictive features contain diverse information that helps a model generalize to out-of-distribution data. We use these features as confounder representations and use them via methods motivated by causal theory to remove bias from models.
  arXiv  Detail & Related papers  (2023-11-28T16:46:14Z)
• Multiple Robust Learning for Recommendation [13.06593469196849]
  In recommender systems, a common problem is the presence of various biases in the collected data. We propose a multiple robust (MR) estimator that can take the advantage of multiple candidate imputation and propensity models to achieve unbiasedness.
  arXiv  Detail & Related papers  (2022-07-09T13:15:56Z)

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.