Cover Image

Learning a Pessimistic Reward Model in RLHF

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Xu, Yinglun, Kang, Hangoo, Suresh, Tarun, Wan, Yuxuan, Singh, Gagandeep
Format:	Preprint
Published:	2025
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

SynCode: LLM Generation with Grammar Augmentation
by: Ugare, Shubham, et al.
Published: (2024)

Binary Reward Labeling: Bridging Offline Preference and Reward-Based Reinforcement Learning
by: Xu, Yinglun, et al.
Published: (2024)

Robust Thompson Sampling Algorithms Against Reward Poisoning Attacks
by: Xu, Yinglun, et al.
Published: (2024)

Universal Black-Box Reward Poisoning Attack against Offline Reinforcement Learning
by: Xu, Yinglun, et al.
Published: (2024)

Is The Watermarking Of LLM-Generated Code Robust?
by: Suresh, Tarun, et al.
Published: (2024)

Stochastic Monkeys at Play: Random Augmentations Cheaply Break LLM Safety Alignment
by: Vega, Jason, et al.
Published: (2024)

CRANE: Reasoning with constrained LLM generation
by: Banerjee, Debangshu, et al.
Published: (2025)

Factored Causal Representation Learning for Robust Reward Modeling in RLHF
by: Yang, Yupei, et al.
Published: (2026)

DINGO: Constrained Inference for Diffusion LLMs
by: Suresh, Tarun, et al.
Published: (2025)

IterGen: Iterative Semantic-aware Structured LLM Generation with Backtracking
by: Ugare, Shubham, et al.
Published: (2024)

Incremental Randomized Smoothing Certification
by: Ugare, Shubham, et al.
Published: (2023)

Optimal Design for Reward Modeling in RLHF
by: Scheid, Antoine, et al.
Published: (2024)

Pessimistic Risk-Aware Policy Learning in Contextual Bandits
by: Wan, Yilong, et al.
Published: (2026)

Two-Step Offline Preference-Based Reinforcement Learning with Constrained Actions
by: Xu, Yinglun, et al.
Published: (2023)

RLHF Workflow: From Reward Modeling to Online RLHF
by: Dong, Hanze, et al.
Published: (2024)

TRAP: Targeted Redirecting of Agentic Preferences
by: Kang, Hangoo, et al.
Published: (2025)

Policy Filtration for RLHF to Mitigate Noise in Reward Models
by: Zhang, Chuheng, et al.
Published: (2024)

Information-Theoretic Reward Modeling for Stable RLHF: Detecting and Mitigating Reward Hacking
by: Miao, Yuchun, et al.
Published: (2025)

Reward Model Overoptimisation in Iterated RLHF
by: Wolf, Lorenz, et al.
Published: (2025)

How to Evaluate Reward Models for RLHF
by: Frick, Evan, et al.
Published: (2024)

Compression Aware Certified Training
by: Xu, Changming, et al.
Published: (2025)

Provably Efficient Online RLHF with One-Pass Reward Modeling
by: Li, Long-Fei, et al.
Published: (2025)

Interactive Machine Learning: From Theory to Scale
by: Zhu, Yinglun
Published: (2025)

Quantile Regression for Distributional Reward Models in RLHF
by: Dorka, Nicolai
Published: (2024)

Reward Shaping to Mitigate Reward Hacking in RLHF
by: Fu, Jiayi, et al.
Published: (2025)

A Unified Pairwise Framework for RLHF: Bridging Generative Reward Modeling and Policy Optimization
by: Xu, Wenyuan, et al.
Published: (2025)

Reward-Robust RLHF in LLMs
by: Yan, Yuzi, et al.
Published: (2024)

Information-Theoretic Reward Decomposition for Generalizable RLHF
by: Mao, Liyuan, et al.
Published: (2025)

Optimistic Model Rollouts for Pessimistic Offline Policy Optimization
by: Zhai, Yuanzhao, et al.
Published: (2024)

Noisy Data is Destructive to Reinforcement Learning with Verifiable Rewards
by: Zhu, Yuxuan, et al.
Published: (2026)

Efficient Active Learning with Abstention
by: Zhu, Yinglun, et al.
Published: (2022)

Cross-Input Certified Training for Universal Perturbations
by: Xu, Changming, et al.
Published: (2024)

Bias Fitting to Mitigate Length Bias of Reward Model in RLHF
by: Zhao, Kangwen, et al.
Published: (2025)

InfoRM: Mitigating Reward Hacking in RLHF via Information-Theoretic Reward Modeling
by: Miao, Yuchun, et al.
Published: (2024)

Active Learning with Neural Networks: Insights from Nonparametric Statistics
by: Zhu, Yinglun, et al.
Published: (2022)

BadReward: Clean-Label Poisoning of Reward Models in Text-to-Image RLHF
by: Duan, Kaiwen, et al.
Published: (2025)

Hyperparameter Tuning Through Pessimistic Bilevel Optimization
by: Ustun, Meltem Apaydin, et al.
Published: (2024)

RLHF in an SFT Way: From Optimal Solution to Reward-Weighted Alignment
by: Du, Yuhao, et al.
Published: (2025)

Towards Multimodal Active Learning: Efficient Learning with Limited Paired Data
by: Zhang, Jiancheng, et al.
Published: (2025)

It Takes Two: On the Seamlessness between Reward and Policy Model in RLHF
by: Lu, Taiming, et al.
Published: (2024)