paper-with-me

홈 › Papers

ResponseRank: Data-Efficient Reward Modeling through Preference Strength Learning

2025-12-31 · Timo Kaufmann, Yannick Metz, Daniel Keim, Eyke Hüllermeier arxiv

Binary choices, as often used for reinforcement learning from human feedback (RLHF), convey only the direction of a preference. A person may choose apples over oranges and bananas over grapes, but which preference is stronger? Strength is crucial for decision-making under uncertainty and generalization of preference models, but hard to measure reliably. Metadata such as response times and inter-annotator agreement can serve as proxies for strength, but are often noisy and confounded. We propose ResponseRank to address the challenge of learning from noisy strength signals. Our method uses relative differences in proxy signals to rank responses to pairwise comparisons by their inferred preference strength. To control for systemic variation, we compare signals only locally within carefully constructed strata. This enables robust learning of utility differences consistent with strength-derived rankings while making minimal assumptions about the strength signal. Our contributions are threefold: (1) ResponseRank, a novel method that robustly learns preference strength by leveraging locally valid relative strength signals; (2) empirical evidence of improved sample efficiency and robustness across diverse tasks: synthetic preference learning (with simulated response times), language modeling (with annotator agreement), and RL control tasks (with simulated episode returns); and (3) the Pearson Distance Correlation (PDC), a novel metric that isolates cardinal utility learning from ordinal accuracy.

📄 PDF Abstract BibTeX arXiv:2512.25023

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

ImplicitRM: Unbiased Reward Modeling from Implicit Preference Data for LLM alignment

2026-03-24 · Hao Wang, Haocheng Yang, Licheng Pan, Lei Shen 외 arxiv

Reward modeling represents a long-standing challenge in reinforcement learning from human feedback (RLHF) for aligning language models. Current reward modeling is heavily contingent upon experimental feedback data with h…

Reinforcement Learning

Two Minds Better Than One: Collaborative Reward Modeling for LLM Alignment

2025-05-15 · Jiazheng Zhang, Wenqing Jing, Zizhuo Zhang, Zhiheng Xi 외

Reward models (RMs) play a pivotal role in aligning large language models (LLMs) with human values. However, noisy preferences in human feedback can lead to reward misgeneralization - a phenomenon where reward models lea…

IRPM: Intergroup Relative Preference Modeling for Pointwise Generative Reward Models

2026-01-02 · Haonan Song, Qingchen Xie, Huan Zhu, Feng Xiao 외 arxiv

Generative Reward Models (GRMs) have demonstrated strong performance in reward modeling, due to their interpretability and potential for refinement through reinforcement learning (RL). However, widely used pairwise GRMs …

Reinforcement Learning

Uncertainty-aware Reward Model: Teaching Reward Models to Know What is Unknown

2024-10-01 · Xingzhou Lou, Dong Yan, Wei Shen, Yuzi Yan 외

Reward models (RM) play a critical role in aligning generations of large language models (LLM) to human expectations. However, prevailing RMs fail to capture the stochasticity within human preferences and cannot effectiv…

Uncertainty Quantification

Omni-Reward: Towards Generalist Omni-Modal Reward Modeling with Free-Form Preferences

2025-10-27 · Zhuoran Jin, Hongbang Yuan, Kejian Zhu, Jiachun Li 외 arxiv

Reward models (RMs) play a critical role in aligning AI behaviors with human preferences, yet they face two fundamental challenges: (1) Modality Imbalance, where most RMs are mainly focused on text and image modalities, …