paper-with-me

Papers

Solving the Inverse Alignment Problem for Efficient RLHF

2024-12-13 · Shambhavi Krishna, Aishwarya Sahoo

Collecting high-quality preference datasets for reinforcement learning from human feedback (RLHF) is resource-intensive and challenging. As a result, researchers often train reward models on extensive offline datasets which aggregate diverse generation sources and scoring/alignment policies. We hypothesize that this aggregation has an averaging effect on reward model scores, which limits signal and impairs the alignment process. Inspired by the field of inverse RL, we define the 'inverse alignment problem' in language model training, where our objective is to optimize the critic's reward for a fixed actor and a fixed offline preference dataset. We hypothesize that solving the inverse alignment problem will improve reward model quality by providing clearer feedback on the policy's current behavior. To that end, we investigate whether repeatedly fine-tuning a reward model on subsets of the offline preference dataset aligned with a periodically frozen policy during RLHF improves upon vanilla RLHF. Our empirical results demonstrate that this approach facilitates superior alignment and faster convergence compared to using an unaligned or out-of-distribution reward model relative to the LLM policy.

📄 PDF Abstract BibTeX arXiv:2412.10529

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage Modelling

Similar Papers 제목 키워드 기반

IR$^3$: Contrastive Inverse Reinforcement Learning for Interpretable Detection and Mitigation of Reward Hacking

2026-02-23 · Mohammad Beigi, Ming Jin, Junshan Zhang, Jiaxin Zhang 외 arxiv

Reinforcement Learning from Human Feedback (RLHF) enables powerful LLM alignment but can introduce reward hacking - models exploit spurious correlations in proxy rewards without genuine alignment. Compounding this, the o…

Reinforcement Learning

PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback

2023-08-03 · Souradip Chakraborty, Amrit Singh Bedi, Alec Koppel, Dinesh Manocha 외

We present a novel unified bilevel optimization-based framework, \textsf{PARL}, formulated to address the recently highlighted critical issue of policy alignment in reinforcement learning using utility or preference-base…

Bilevel OptimizationProcedure Learningreinforcement-learningReinforcement Learning (RL)

Learning from Failures: Understanding LLM Alignment through Failure-Aware Inverse RL

2025-10-07 · Nyal Patel, Matthieu Bou, Arjun Jagota, Satyapriya Krishna 외 arxiv

Reinforcement Learning from Human Feedback (RLHF) aligns Large Language Models (LLMs) with human preferences, yet the underlying reward signals they internalize remain hidden, posing a critical challenge for interpretabi…

Reinforcement Learning

Principled Penalty-based Methods for Bilevel Reinforcement Learning and RLHF

2024-02-10 · Han Shen, Zhuoran Yang, Tianyi Chen

Bilevel optimization has been recently applied to many machine learning tasks. However, their applications have been restricted to the supervised learning setting, where static objective functions with benign structures …

Bilevel Optimizationreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Insights from the Inverse: Reconstructing LLM Training Goals Through Inverse RL

2024-10-16 · Jared Joselowitz, Arjun Jagota, Satyapriya Krishna, Sonali Parbhoo

Large language models (LLMs) trained with Reinforcement Learning from Human Feedback (RLHF) have demonstrated remarkable capabilities, but their underlying reward functions and decision-making processes remain opaque. Th…

Decision Makingreinforcement-learningReinforcement Learning