Abstract:Large Language Models (LLMs), particularly slow-thinking models, often exhibit severe hallucination, outputting incorrect content due to an inability to accurately recognize knowledge boundaries during reasoning. While Reinforcement Learning (RL) can enhance complex reasoning abilities, its outcome-oriented reward mechanism often lacks factual supervision over the thinking process, further exacerbating the hallucination problem. To address the high hallucination in slow-thinking models, we propose Knowledge-enhanced RL, KnowRL. KnowRL guides models to perform fact-based slow thinking by integrating a factuality reward, based on knowledge verification, into the RL training process, helping them recognize their knowledge boundaries. KnowRL guides models to perform fact-based slow thinking by integrating a factuality reward, based on knowledge verification, into the RL training process, helping them recognize their knowledge boundaries. This targeted factual input during RL training enables the model to learn and internalize fact-based reasoning strategies. By directly rewarding adherence to facts within the reasoning steps, KnowRL fosters a more reliable thinking process. Experimental results on three hallucination evaluation datasets and two reasoning evaluation datasets demonstrate that KnowRL effectively mitigates hallucinations in slow-thinking models while maintaining their original strong reasoning capabilities. Our code is available at this https URL.
| Comments: | ACL 2026 |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2506.19807 [cs.AI] |
| (or arXiv:2506.19807v4 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2506.19807 arXiv-issued DOI via DataCite |
From: Ningyu Zhang [view email]
[v1]
Tue, 24 Jun 2025 17:17:17 UTC (8,504 KB)
[v2]
Sun, 6 Jul 2025 16:11:23 UTC (8,507 KB)
[v3]
Wed, 8 Oct 2025 16:56:59 UTC (7,524 KB)
[v4]
Thu, 16 Apr 2026 16:50:02 UTC (8,707 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。