# Learning to Reason for Hallucination Span Detection

arXiv: https://arxiv.org/abs/2510.02173 · HTML: https://mlsanity.com/paper/2510.02173/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Hallucination detection and mitigation, one of 17 directions in LLM reasoning.
- Builds on 20 of its 33 references that are in the map.
- First posted on arXiv 2 Oct 2025.

## Placed in

- LLM reasoning / [Hallucination detection and mitigation](https://mlsanity.com/lr/hallucination-detection-and-mitigation/)

## Builds on (most cited first, 20 of 20)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [Understanding R1-Zero-Like Training: A Critical Perspective](https://mlsanity.com/paper/2503.20783/) (2025)
- [Reasoning with Language Model is Planning with World Model](https://mlsanity.com/paper/2305.14992/) (2023)
- [ReTool: Reinforcement Learning for Strategic Tool Use in LLMs](https://mlsanity.com/paper/2504.11536/) (2025)
- [Sentiment Analysis in the Era of Large Language Models: A Reality Check](https://mlsanity.com/paper/2305.15005/) (2023)
- [Why Language Models Hallucinate](https://mlsanity.com/paper/2509.04664/) (2025)
- [LLM Internal States Reveal Hallucination Risk Faced With a Query](https://mlsanity.com/paper/2407.03282/) (2024)
- [ChatGPT as a Factual Inconsistency Evaluator for Text Summarization](https://mlsanity.com/paper/2303.15621/) (2023)
- [Lynx: An Open Source Hallucination Evaluation Model](https://mlsanity.com/paper/2407.08488/) (2024)
- [rStar2-Agent: Agentic Reasoning Technical Report](https://mlsanity.com/paper/2508.20722/) (2025)
- [HaluEval-Wild: Evaluating Hallucinations of Language Models in the Wild](https://mlsanity.com/paper/2403.04307/) (2024)
- [Optimizing Safe and Aligned Language Generation: A Multi-Objective GRPO Approach](https://mlsanity.com/paper/2503.21819/) (2025)
- [R1-Code-Interpreter: LLMs Reason with Code via Supervised and Multi-stage Reinforcement Learning](https://mlsanity.com/paper/2505.21668/) (2025)
- [Improving Generalization in Intent Detection: GRPO with Reward-Based Curriculum Sampling](https://mlsanity.com/paper/2504.13592/) (2025)
- [CLATTER: Comprehensive Entailment Reasoning for Hallucination Detection](https://mlsanity.com/paper/2506.05243/) (2025)
- [Hallucinated Span Detection with Multi-View Attention Features](https://mlsanity.com/paper/2504.04335/) (2025)

## Built on by (most cited first, 5 of 5)

- [DOCR-Inspector: Fine-Grained and Automated Evaluation of Document Parsing with VLM](https://mlsanity.com/paper/2512.10619/) (2025)
- [Beyond Document Grounding: Span-Level Hallucination Detection over Code, Tool Output, and Documents](https://mlsanity.com/paper/2607.00895/) (2026)
- [TPA: Next Token Probability Attribution for Detecting Hallucinations in RAG](https://mlsanity.com/paper/2512.07515/) (2025)
- [Hallucination Self-Play: Bootstrapping Reinforced Detector via Evolved Generator](https://mlsanity.com/paper/2607.07993/) (2026)
- [Hallucination Span Detection with Input-Side Evidence Alignment](https://mlsanity.com/paper/2608.15804/) (2026)

## Abstract (arXiv)

Large language models (LLMs) often generate hallucinations -- unsupported content that undermines reliability. While most prior works frame hallucination detection as a binary task, many real-world applications require identifying hallucinated spans, which is a multi-step decision making process. This naturally raises the question of whether explicit reasoning can help the complex task of detecting hallucination spans. To answer this question, we first evaluate pretrained models with and without Chain-of-Thought (CoT) reasoning, and show that CoT reasoning has the potential to generate at least one correct answer when sampled multiple times. Motivated by this, we propose RL4HS, a reinforcement learning framework that incentivizes reasoning with a span-level reward function. RL4HS builds on Group Relative Policy Optimization and introduces Class-Aware Policy Optimization to mitigate reward imbalance issue. Experiments on the RAGTruth benchmark (summarization, question answering, data-to-text) show that RL4HS surpasses pretrained reasoning models and supervised fine-tuning, demonstrating the necessity of reinforcement learning with span-level rewards for detecting hallucination spans.