mlsanityData to 14 Oct 2026

Learning to Reason for Factuality

arXiv 2508.05618 ↗First posted 7 Aug 2025cs.CLcited by 24

On the map: LLM reasoning → Hallucination detection and mitigation

Abstract (arXiv)

Reasoning Large Language Models (R-LLMs) have significantly advanced complex reasoning tasks but often struggle with factuality, generating substantially more hallucinations than their non-reasoning counterparts on long-form factuality benchmarks. However, extending online Reinforcement Learning (RL), a key component in recent R-LLM advancements, to the long-form factuality setting poses several unique challenges due to the lack of reliable verification methods. Previous work has utilized automatic factuality evaluation frameworks such as FActScore to curate preference data in the offline RL setting, yet we find that directly leveraging such methods as the reward in online RL leads to reward hacking in multiple ways, such as producing less detailed or relevant responses. We propose a novel reward function that simultaneously considers the factual precision, response detail level, and answer relevance, and applies online RL to learn high quality factual reasoning. Evaluated on six long-form factuality benchmarks, our factual reasoning model achieves an average reduction of 23.1 percentage points in hallucination rate, a 23% increase in answer detail level, and no degradation in the overall response helpfulness.

Key facts. Cited by 24 papers in the map (data through 14 Oct 2026). Placed in Hallucination detection and mitigation, one of 17 directions in LLM reasoning. Builds on 19 of its 42 references that are in the map. First posted on arXiv 7 Aug 2025.

Builds on

19 of 42 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025KnowRL: Exploring Knowledgeable Reinforcement Learning for Factuality24 Jun
  2. 2025Reasoning Models Hallucinate More: Factuality-Aware Reinforcement Learning for Large Reasoning Models30 May

Other references · 17

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  3. 2024Qwen2.5 Technical Reportcited by 4,942
  4. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,444
  5. 2025Understanding R1-Zero-Like Training: A Critical Perspectivecited by 1,345
  6. 2025s1: Simple test-time scalingcited by 1,341
  7. 2024OpenAI o1 System Cardcited by 1,244
  8. 2024Tulu 3: Pushing Frontiers in Open Language Model Post-Trainingcited by 1,053
  9. 2024DeepSeek-V3 Technical Reportcited by 811
  10. 2023AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedbackcited by 748
  11. 2025LIMO: Less is More for Reasoningcited by 498
  12. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Modelscited by 384
5 more
  1. 2024ReFT: Reasoning with Reinforced Fine-Tuningcited by 239
  2. 2025NaturalThoughts: Selecting and Distilling Reasoning Traces for General Reasoning Taskscited by 32
  3. 2024FactBench: A Dynamic Benchmark for In-the-Wild Language Model Factuality Evaluationcited by 25
  4. 2025Bridging Offline and Online Reinforcement Learning for LLMscited by 25
  5. 2025FACTORY: A Challenging Human-Verified Prompt Set for Long-Form Factualitycited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 9; 2026: 15.

24 papers.

Sort:

By area:

  1. 2025TruthRL: Incentivizing Truthful LLMs via Reinforcement Learningcited by 27
  2. 2025Position: The Hidden Costs and Measurement Gaps of Reinforcement Learning with Verifiable Rewardscited by 12
  3. 2026Baichuan-M3: Modeling Clinical Inquiry for Reliable Medical Decision-Makingcited by 11
  4. 2025Enhancing Large Language Model Reasoning with Reward Models: An Analytical Surveycited by 8
  5. 2026Self-Improving Pretraining: using post-trained models to pretrain better modelscited by 6
  6. 2025KnowRL: Teaching Language Models to Know What They Knowcited by 5
  7. 2025Train for Truth, Keep the Skills: Binary Retrieval-Augmented Reward Mitigates Hallucinationscited by 5
  8. 2026Scaling Large Reasoning Models beyond Human Supervision: A Path toward Superintelligencecited by 5
  9. 2025Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factualitycited by 4
  10. 2025MR-Align: Meta-Reasoning Informed Factuality Alignment for Large Reasoning Modelscited by 4
  11. 2026FaithRL: Learning to Reason Faithfully through Step-Level Faithfulness Maximizationcited by 4
  12. 2026Beyond Reasoning: Reinforcement Learning Unlocks Parametric Knowledge in LLMscited by 4
12 more
  1. 2026Beyond Math and Code: Lightweight Corpus-Grounded Process Rewards for Factual Question Answeringcited by 4
  2. 2026Mitigating Factual Hallucination in Large Reasoning Models via Mixed-Mode Advantage Regularizationcited by 3
  3. 2026Think Through Uncertainty: Improving Long-Form Generation Factuality via Reasoning Calibrationcited by 2
  4. 2026From Refuse to Richness: Rubric Rewards for Long-Form Hallucination Reinforcement Learningcited by 1
  5. 2025From Proof to Program: Characterizing Tool-Induced Reasoning Hallucinations in Large Language Modelscited by 0
  6. 2025Enhancing Reliability across Short and Long-Form QA via Reinforcement Learningcited by 0
  7. 2026Improving Latent Generalization Using Test-time Computecited by 0
  8. 2026Only Say What You Know: Calibration-Aware Generation for Long-Form Factualitycited by 0
  9. 2026BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigationcited by 0
  10. 2026From Storage to Access: Verifiable Activation of Parametric Knowledge in LLMs via Explicit Priming and Implicit Reasoningcited by 0
  11. 2026Hallucination Across the Reasoning Lifecycle: Interface Visibility, Causal Evidence, and Release Control in Large Reasoning Modelscited by 0
  12. 2026Copying Is Not Grounding: Training Faithful Long-Form RAG with Authored-Sentence Creditcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Trustworthy Reasoning: Evaluating and Enhancing Factual Accuracy in LLM Intermediate Thought Processes
  2. 2025The Climb Carves Wisdom Deeper Than the Summit: On the Noisy Rewards in Learning to Reason
  3. 2025Rethinking Reasoning Quality in Large Language Models through Enhanced Chain-of-Thought via RL
  4. 2025Beyond Correctness: Confidence-Aware Reward Modeling for Enhancing Large Language Model Reasoning
  5. 2024FLAME: Factuality-Aware Alignment for Large Language Models
  6. 2024Rejection Improves Reliability: Training LLMs to Refuse Unknown Questions Using RL from Knowledge Feedback
  7. 2025Thinking, Faithful and Stable: Mitigating Hallucinations in LLMs
  8. 2025Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs