mlsanityData to 7 Oct 2026

Evidence-State Rewards for Long-Context Reasoning

arXiv 2607.02073 ↗First posted 2 Jul 2026cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Long-context reasoning requires models to locate, revise, and synthesize evidence distributed across lengthy inputs. Existing long-context RL methods usually reward final answers or static evidence extraction, offering little feedback on how intermediate actions change the model's evidence state. We propose Maven, a reinforcement learning framework with an editable evidence memory. Maven defines an answer-conditioned evidence-state value and rewards action-level state transitions: add actions are credited by marginal gain and hindsight contribution, link actions by evidence synergy, and drop actions by improved answer support after removing misleading evidence. These rewards are assigned to the corresponding action spans in GRPO. Across Llama and Qwen models on LongBench v2, LongReason, and RULER, Maven outperforms outcome-only RL and evidence-identification baselines, producing more sufficient evidence sets and lower distractor retention. Our results show that long-context RL benefits from optimizing stateful evidence navigation rather than one-shot evidence extraction.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 24 of its 28 references that are in the map. First posted on arXiv 2 Jul 2026.

Builds on

24 of 28 references.

References · 24

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  2. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  3. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  4. 2023Let's Verify Step by Stepcited by 3,598
  5. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  6. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
  7. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,371
  8. 2024RULER: What's the Real Context Size of Your Long-Context Language Models?cited by 1,270
  9. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,259
  10. 2023Extending Context Window of Large Language Models via Positional Interpolationcited by 721
  11. 2025The Lessons of Developing Process Reward Models in Mathematical Reasoningcited by 402
  12. 2024LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitaskscited by 367
12 more
  1. 2024LongRoPE: Extending LLM Context Window Beyond 2 Million Tokenscited by 302
  2. 2024Chain of Agents: Large Language Models Collaborating on Long-Context Taskscited by 201
  3. 2025Process Reward Models That Thinkcited by 118
  4. 2024LongRAG: Enhancing Retrieval-Augmented Generation with Long-context LLMscited by 104
  5. 2024LongRAG: A Dual-Perspective Retrieval-Augmented Generation Paradigm for Long-Context Question Answeringcited by 49
  6. 2025LongReason: A Synthetic Long-Context Reasoning Benchmark via Context Expansioncited by 47
  7. 2025QwenLong-L1: Towards Long-Context Large Reasoning Models with Reinforcement Learningcited by 46
  8. 2024LongReward: Improving Long-context Large Language Models with AI Feedbackcited by 34
  9. 2025LoongRL: Reinforcement Learning for Advanced Reasoning over Long Contextscited by 25
  10. 2026LongRLVR: Long-Context Reinforcement Learning Requires Verifiable Context Rewardscited by 13
  11. 2026LongR: Unleashing Long-Context Reasoning via Reinforcement Learning with Dense Utility Rewardscited by 9
  12. 2026Evidence-Augmented Policy Optimization with Reward Co-Evolution for Long-Context Reasoningcited by 7

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026CoEM: Empowering Long-Context Reasoning with Commit-on-Evidence Memory
  2. 2026Copy Less, Ground More: Overcoming Repetitive Copying in Long-Context Reasoning via Evidence-Aware Reinforcement Learning
  3. 2026ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
  4. 2026Beyond Reward Engineering: A Data Recipe for Long-Context Reinforcement Learning
  5. 2026Learning Evidence Highlighting for Frozen LLMs
  6. 2026LatentHarness: Learning Latent Actions for Memory and Reasoning via Counterfactual Policy Distillation
  7. 2026A Decomposition Perspective to Long-context Reasoning for LLMs
  8. 2025Look Back to Reason Forward: Revisitable Memory for Long-Context LLM Agents