Learning to Extract Rational Evidence via Reinforcement Learning for Retrieval-Augmented Generation

ACL ARR 2026 January Submission3452 Authors

Keywords: Retreival-Augmented Genetaion, Evidence Extraction, LLMs

Abstract: Retrieval-Augmented Generation (RAG) effectively improves the accuracy of Large Language Models (LLMs). However, retrieval noises significantly undermine the quality of LLMs' generation, necessitating the development of denoising mechanisms. Previous works extract evidence straightforwardly without deep thinking, which may risk filtering out key clues and struggle with generalization. To this end, we propose EviOmni, which learns to extract rational evidence via reasoning first and then extracting. Specifically, EviOmni integrates evidence reasoning and evidence extraction into one unified trajectory, followed by knowledge token masking to avoid information leakage, optimized via on-policy reinforcement learning with verifiable rewards in terms of answer, length, and format. Extensive experiments on five benchmark datasets show the superiority of EviOmni, which provides compact and high-quality evidence, enhances the accuracy of downstream tasks, and supports both traditional and agentic RAG systems.

Paper Type: Long

Research Area: Retrieval-Augmented Language Models

Research Area Keywords: Generation,Question Answering,Information Retrieval and Text Mining

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Approaches to low-resource settings, Approaches low compute settings-efficiency

Languages Studied: English

Submission Number: 3452