mlsanityData to 14 Oct 2026

Hybrid Reasoning Agents

arXiv 2610.18861 ↗First posted 13 Oct 2026cs.CL cs.AI cs.LGcited by 0

On the map: LLM reasoning → Reasoning for agents, planning and code

also in 1 more
Abstract (arXiv)

How can we build an agent that thinks beyond human language, yet still acts through it in the external environment? Modern LLM agents still reason and interact through text, while latent reasoning offers an efficient internal medium but lacks native tool access and reasoning transparency. We bridge these two directions with Hybrid Reasoning Agents (HybridAgents), which dynamically navigate between latent reasoning and text-based interaction throughout a unified generation process. Architecturally, a lightweight HybridLink enables seamless transitions between text and latent modes, while the agent's continuous working memory preserves lossless hybrid context across multiple turns. These designs preserve tool-integrated agentic capabilities while making latent thoughts directly inspectable. Building on this architecture, we further develop dedicated post-training algorithms for hybrid reasoning: HybridSFT stabilizes tool-integrated hybrid generation by teaching the backbone to coordinate with HybridLink, then HybridRL scales reasoning by learning when to think in latent space and when to act through text. We evaluate HybridAgent at 3 model scales across 11 benchmarks covering mathematics, science, medicine, search, and agentic reasoning. Compared with frontier LLM Agents and agentic frameworks spanning both text-based and latent methods, HybridAgent consistently improves performance by 10.4%, while delivering 35.5%-54.7% token reduction, 1.8x-2.6x faster inference, and 2.7x-3.7x cost savings. Code and data are available at https://github.com/togethercomputer/HybridAgent.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 83 of its 99 references that are in the map. First posted on arXiv 13 Oct 2026.

Builds on

83 of 99 references.

Closest ancestors · 3

Same direction, newest first

  1. 2026Stealing Reasoning Traces from Proprietary LLM APIs10 Aug
  2. 2025Tool-Star: Empowering LLM-Brained Multi-Tool Reasoner via Reinforcement Learning22 May
  3. 2025ReTool: Reinforcement Learning for Strategic Tool Use in LLMs15 Apr

Other references · 80

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2025Qwen3 Technical Reportcited by 8,820
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  5. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  6. 2024GPT-4o System Cardcited by 4,588
  7. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,879
  8. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  9. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 2,057
  10. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,947
  11. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,444
  12. 2023SGLang: Efficient Execution of Structured Language Model Programscited by 1,141
38 more
  1. 2023GAIA: a benchmark for General AI Assistantscited by 1,128
  2. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 1,017
  3. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 916
  4. 2015Scheduled Sampling for Sequence Prediction with Recurrent Neural Networkscited by 803
  5. 2024Training Large Language Models to Reason in a Continuous Latent Spacecited by 797
  6. 2015Sequence Level Training with Recurrent Neural Networkscited by 576
  7. 2019ERASER: A Benchmark to Evaluate Rationalized NLP Modelscited by 530
  8. 2025Humanity's Last Examcited by 491
  9. 2025Search-o1: Agentic Search-Enhanced Large Reasoning Modelscited by 433
  10. 2023Cognitive Architectures for Language Agentscited by 429
  11. 2025Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approachcited by 415
  12. 2025Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscationcited by 384
  13. 2025ToolRL: Reward is All Tool Learning Needscited by 362
  14. 2025Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agentscited by 356
  15. 2025WebThinker: Empowering Large Reasoning Models with Deep Research Capabilitycited by 335
  16. 2025Chain of Draft: Thinking Faster by Writing Lesscited by 297
  17. 2024ProcessBench: Identifying Process Errors in Mathematical Reasoningcited by 292
  18. 2025DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environmentscited by 286
  19. 2023Prompt Cache: Modular Attention Reuse for Low-Latency Inferencecited by 268
  20. 2025Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safetycited by 249
  21. 2025Reasoning with Latent Thoughts: On the Power of Looped Transformerscited by 240
  22. 2024Tool Learning with Large Language Models: A Surveycited by 223
  23. 2024Compressed Chain of Thought: Efficient Reasoning Through Dense Representationscited by 201
  24. 2025WebWalker: Benchmarking LLMs in Web Traversalcited by 175
  25. 2025Agentic Reinforced Policy Optimizationcited by 172
  26. 2025Measuring AI Ability to Complete Long Software Taskscited by 156
  27. 2025The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Taskscited by 152
  28. 2026Kimi K3: Open Frontier Intelligencecited by 138
  29. 2025Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Spacecited by 117
  30. 2025SoftCoT: Soft Chain-of-Thought for Efficient Reasoning with LLMscited by 116
  31. 2025MemGen: Weaving Generative Latent Memory for Self-Evolving Agentscited by 109
  32. 2025Think Silently, Think Fast: Dynamic Latent Compression of LLM Reasoning Chainscited by 85
  33. 2024ChunkAttention: Efficient Self-Attention with Prefix-Aware KV Cache and Two-Phase Partitioncited by 82
  34. 2025Monet: Reasoning in Latent Visual Space Beyond Images and Languagecited by 82
  35. 2025A Survey on Latent Reasoningcited by 81
  36. 2025Reasoning Beyond Language: A Comprehensive Survey on Latent Chain-of-Thought Reasoningcited by 78
  37. 2025Reasoning by Superposition: A Theoretical Perspective on Chain of Continuous Thoughtcited by 77
  38. 2026DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compressioncited by 76

50 most cited of 80 All 80 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Adaptive Latent Agentic Reasoning
  2. 2025Latent Collaboration in Multi-Agent Systems
  3. 2025Unleashing Diverse Thinking Modes in LLMs through Multi-Agent Collaboration
  4. 2025A2FM: An Adaptive Agent Foundation Model for Tool-Aware Hybrid Reasoning
  5. 2025Revisiting Multi-Agent Debate as Test-Time Scaling: A Systematic Study of Conditional Effectiveness
  6. 2024Textualized Agent-Style Reasoning for Complex Tasks by Multiple Round LLM Generation
  7. 2025Agentic Reasoning: A Streamlined Framework for Enhancing LLM Reasoning with Agentic Tools
  8. 2026A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agents