# LongHarness: Advancing Agentic Reasoning over 100M-Token Corpora via Stateful Evidence Management arXiv: https://arxiv.org/abs/2610.13075 ยท HTML: https://mlsanity.com/paper/2610.13075/ ## Key facts - No citing papers in the map yet (data through 14 Oct 2026). - Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use. - Builds on 38 of its 51 references that are in the map. - First posted on arXiv 9 Oct 2026. ## Placed in - LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/) ## Builds on (most cited first, 30 of 38) - [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022) - [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023) - [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024) - [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024) - [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026) - [LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks](https://mlsanity.com/paper/2412.15204/) (2024) - [Chain of Agents: Large Language Models Collaborating on Long-Context Tasks](https://mlsanity.com/paper/2406.02818/) (2024) - [Qwen2.5-1M Technical Report](https://mlsanity.com/paper/2501.15383/) (2025) - [A Human-Inspired Reading Agent with Gist Memory of Very Long Contexts](https://mlsanity.com/paper/2402.09727/) (2024) - [Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses](https://mlsanity.com/paper/2604.25850/) (2026) - [Walking Down the Memory Maze: Beyond Context Limit through Interactive Reading](https://mlsanity.com/paper/2310.05029/) (2023) - [Recursive Language Models](https://mlsanity.com/paper/2512.24601/) (2025) - [LongReason: A Synthetic Long-Context Reasoning Benchmark via Context Expansion](https://mlsanity.com/paper/2501.15089/) (2025) - [QwenLong-L1: Towards Long-Context Large Reasoning Models with Reinforcement Learning](https://mlsanity.com/paper/2505.17667/) (2025) - [GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models](https://mlsanity.com/paper/2406.14550/) (2024) - [LoongRL: Reinforcement Learning for Advanced Reasoning over Long Contexts](https://mlsanity.com/paper/2510.19363/) (2025) - [QwenLong-L1.5: Post-Training Recipe for Long-Context Reasoning and Memory Management](https://mlsanity.com/paper/2512.12967/) (2025) - [LongRoPE2: Near-Lossless LLM Context Window Scaling](https://mlsanity.com/paper/2502.20082/) (2025) - [Beyond the Final Answer: Evaluating the Reasoning Trajectories of Tool-Augmented Agents](https://mlsanity.com/paper/2510.02837/) (2025) - [LongRLVR: Long-Context Reinforcement Learning Requires Verifiable Context Rewards](https://mlsanity.com/paper/2603.02146/) (2026) - [Prime Agent: A Self-Improving RLM Harness](https://mlsanity.com/paper/2608.23552/) (2026) - [Coding Agents are Effective Long-Context Processors](https://mlsanity.com/paper/2603.20432/) (2026) - [Self-Taught Agentic Long Context Understanding](https://mlsanity.com/paper/2502.15920/) (2025) - [LongR: Unleashing Long-Context Reasoning via Reinforcement Learning with Dense Utility Rewards](https://mlsanity.com/paper/2602.05758/) (2026) - [Memory-R2: Fair Credit Assignment for Long-Horizon Memory-Augmented LLM Agents](https://mlsanity.com/paper/2605.21768/) (2026) - [Recursive Experiential-Working Memory Evolution for Long-Horizon Agent Harnesses](https://mlsanity.com/paper/2608.24876/) (2026) - [Learning to Remember: End-to-End Training of Memory Agents for Long-Context Reasoning](https://mlsanity.com/paper/2602.18493/) (2026) - [GoLongRL: Capability-Oriented Long Context Reinforcement Learning with Multitask Alignment](https://mlsanity.com/paper/2605.19577/) (2026) - [LEDGERMIND: Provenance-Constrained Multimodal Agentic Reasoning with a Structured Evidence Ledger](https://mlsanity.com/paper/2607.28374/) (2026) - [DocPuzzle: A Process-Aware Benchmark for Evaluating Realistic Long-Context Reasoning Capabilities](https://mlsanity.com/paper/2502.17807/) (2025) ## Abstract (arXiv) Corpus-level long-context reasoning requires agents to acquire and integrate evidence across document collections that exceed their context windows. Existing agent harnesses enable incremental access to these corpora, but acquired evidence and evolving conclusions often remain entangled in a growing interaction history. This implicit state management can lead to two coupled failure modes: unbounded exploration, where retrieval expands the interaction history without resolving decision-relevant information gaps, and evidence-state drift, where previously acquired evidence or subsequent corrections fail to inform the current conclusion. We present LongHarness, a model-agnostic harness that makes the interaction between acquired evidence and evolving conclusions explicit. Its core mechanism, Evidence-Conclusion Coevolution, uses unresolved conclusions to guide exploration toward missing information, while newly acquired evidence strengthens or revises the conclusions that directed the search. LongHarness couples this process with Evidence Lifecycle Management, which keeps accumulated evidence current and traceable throughout the interaction. Together, these mechanisms focus retrieval on remaining information needs and keep conclusions aligned with current evidence. We evaluate LongHarness on OfficeQA Pro V1 and V2, AA-LCR, and LongBench v2 using GLM-5.3-Flash and Qwen3.8-Flash. Compared with direct reasoning using the same backbone, LongHarness improves GLM-5.3-Flash exact match on OfficeQA Pro from 15.8% to 60.2% on V1 and from 4.4% to 61.1% on V2. Against state-of-the-art agent harnesses, it surpasses Claude Code by 12.0 percentage points on V1 and Codex by 18.9 points on V2. With Qwen3.8-Flash, LongHarness reaches 61.7% on V1, outperforming Claude Code by 8.3 points and Codex by 12.0 points.