mlsanityData to 14 Oct 2026

Reasoning Beyond Language: A Comprehensive Survey on Latent Chain-of-Thought Reasoning

arXiv 2505.16782 ↗First posted 22 May 2025cs.CLcited by 78

On the map: LLM reasoning → Prompting and chain-of-thought reasoning

Abstract (arXiv)

Large Language Models (LLMs) have shown impressive performance on complex tasks through Chain-of-Thought (CoT) reasoning. However, conventional CoT relies on explicitly verbalized intermediate steps, which constrains its broader applicability, particularly in abstract reasoning tasks beyond language. To address this, there has been growing research interest in textitlatent CoT reasoning, where the reasoning process is embedded within latent spaces. By decoupling reasoning from explicit language generation, latent CoT offers the promise of richer cognitive representations and facilitates more flexible, faster inference. This paper aims to present a comprehensive overview of this emerging paradigm and establish a systematic taxonomy. We analyze recent advances in methods, categorizing them from token-wise horizontal approaches to layer-wise vertical strategies. We then provide in-depth discussions of these methods, highlighting their design principles, applications, and remaining challenges. We hope that our survey provides a structured foundation for advancing this promising direction in LLM reasoning. The relevant papers will be regularly updated at https://github.com/EIT-NLP/Awesome-Latent-CoT.

Key facts. Cited by 78 papers in the map (data through 14 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 37 of its 140 references that are in the map. First posted on arXiv 22 May 2025.

Builds on

37 of 140 references.

Closest ancestors · 12

Same direction, newest first

  1. 2025Internal Chain-of-Thought: Empirical Evidence for Layer-wise Subtask Scheduling in LLMs20 May
  2. 2025Reasoning by Superposition: A Theoretical Perspective on Chain of Continuous Thought18 May
  3. 2025Reasoning Models Don't Always Say What They Think8 May
  4. 2025TwT: Thinking without Tokens by Habitual Reasoning Distillation with Multi-Teachers' Guidance31 Mar
  5. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models12 Mar
  6. 2024LLMs Faithfully and Iteratively Compute Answers During CoT: A Systematic Analysis With Multi-step Arithmetics2 Dec
  7. 2024Do LLMs Really Think Step-by-step In Implicit Reasoning?24 Nov
  8. 2024Uncovering Latent Chain of Thought Vectors in Language Models21 Sep
  9. 2024Expediting and Elevating Large Language Model Reasoning via Hidden Chain-of-Thought Decoding13 Sep
  10. 2024Distilling System 2 into System 18 Jul
  11. 2024From Explicit CoT to Implicit CoT: Learning to Internalize CoT Step by Step23 May
  12. 2023Implicit Chain of Thought Reasoning via Knowledge Distillation2 Nov

Other references · 25

  1. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  3. 2024Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Securitycited by 260
  4. 2025Reasoning with Latent Thoughts: On the Power of Looped Transformerscited by 240
  5. 2025A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?cited by 223
  6. 2024Compressed Chain of Thought: Efficient Reasoning Through Dense Representationscited by 201
  7. 2025Efficient Reasoning with Hidden Thinkingcited by 75
  8. 2025SEAL: Steerable Reasoning Calibration of Large Language Models for Freecited by 68
  9. 2025Inner Thinking Transformer: Leveraging Dynamic Depth Scaling to Foster Adaptive Internal Thinkingcited by 57
  10. 2025Think Before Recommend: Unleashing the Latent Reasoning Power for Sequential Recommendationcited by 51
  11. 2025Do Language Models Use Their Depth Efficiently?cited by 50
  12. 2024Multimodal Latent Language Modeling with Next-Token Diffusioncited by 48
13 more
  1. 2025SSR: Enhancing Depth Perception in Vision-Language Models via Rationale-Guided Spatial Reasoningcited by 46
  2. 2024Towards Efficient LLM Grounding for Embodied Multi-Agent Collaborationcited by 45
  3. 2025Reasoning to Learn from Latent Thoughtscited by 45
  4. 2025Seek in the Dark: Reasoning via Test-Time Instance-Level Policy Gradient in Latent Spacecited by 43
  5. 2025SoftCoT++: Test-Time Scaling with Soft Chain-of-Thought Reasoningcited by 41
  6. 2024Thinking Tokens for Language Modelingcited by 24
  7. 2024Distributional reasoning in LLMs: Parallel reasoning processes in multi-hop reasoningcited by 21
  8. 2025Text Generation Beyond Discrete Token Samplingcited by 21
  9. 2025Implicit Reasoning in Transformers is Reasoning through Shortcutscited by 14
  10. 2025Enhancing Latent Computation in Transformers with Latent Tokenscited by 14
  11. 2025Exploring the Necessity of Reasoning in LLM-based Agent Scenarioscited by 10
  12. 2025Learning Refined Document Representations for Dense Retrieval via Deliberate Thinkingcited by 5
  13. 2025Enhancing Non-Core Language Instruction-Following in Speech LLMs via Semi-Implicit Cross-Lingual CoT Reasoningcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 14; 2026: 64.

78 papers.

Sort:

By area:

  1. 2025Hierarchical Reasoning Modelcited by 177
  2. 2025Agentic Reinforced Policy Optimizationcited by 172
  3. 2025Latent Collaboration in Multi-Agent Systemscited by 73
  4. 2025Implicit Reasoning in Large Language Models: A Comprehensive Surveycited by 43
  5. 2026The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlookcited by 43
  6. 2025VisMem: Latent Vision Memory Unlocks Potential of Vision-Language Modelscited by 37
  7. 2026LaST0: Latent Spatio-Temporal Chain-of-Thought for Robotic Vision-Language-Action Modelcited by 31
  8. 2025A Survey of Token Compression for Efficient Multimodal Large Language Modelscited by 25
  9. 2025ReasonMap: Towards Fine-Grained Visual Reasoning from Transit Mapscited by 23
  10. 2025SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMscited by 22
  11. 2025RewardMap: Tackling Sparse Rewards in Fine-grained Visual Reasoning via Multi-Stage Reinforcement Learningcited by 21
  12. 2025SofT-GRPO: Surpassing Discrete-Token LLM Reinforcement Learning via Gumbel-Reparameterized Soft-Thinking Policy Optimizationcited by 21
38 more
  1. 2025Latent Chain-of-Thought World Modeling for End-to-End Drivingcited by 15
  2. 2025Parallel Test-Time Scaling for Latent Reasoning Modelscited by 14
  3. 2026Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformerscited by 13
  4. 2026SpiralFormer: Looped Transformers Can Learn Hierarchical Dependencies via Multi-Resolution Recursioncited by 12
  5. 2026Dynamics Within Latent Chain-of-Thought: An Empirical Study of Causal Structurecited by 11
  6. 2026PLUME: Latent Reasoning Based Universal Multimodal Embeddingcited by 11
  7. 2026Large Reasoning Models Are (Not Yet) Multilingual Latent Reasonerscited by 9
  8. 2026LaSER: Internalizing Explicit Reasoning into Latent Space for Dense Retrievalcited by 9
  9. 2026Are Latent Reasoning Models Easily Interpretable?cited by 9
  10. 2025KaVa: Latent Reasoning via Compressed KV-Cache Distillationcited by 8
  11. 2025MeSH: Memory-as-State-Highways for Recursive Transformerscited by 8
  12. 2026Show, Don't Tell: Morphing Latent Reasoning into Image Generationcited by 8
  13. 2026ThinkRouter: Efficient Reasoning via Routing Thinking between Latent and Discrete Spacescited by 8
  14. 2026LaST-R1: Reinforcing Robotic Manipulation via Adaptive Physical Latent Reasoningcited by 8
  15. 2026ATLAS: Verifier-Guided Adaptive Latent Activation Steering for Efficient LLM Reasoningcited by 6
  16. 2026S2GR: Stepwise Semantic-Guided Reasoning in Latent Space for Generative Recommendationcited by 6
  17. 2026Latent Chain-of-Thought as Planning: Decoupling Reasoning from Verbalizationcited by 6
  18. 2026Towards Explainable Industrial Anomaly Detection via Knowledge-Guided Latent Reasoningcited by 6
  19. 2026Chain Of Thought Compression: A Theoretical Analysiscited by 5
  20. 2026Step-resolved data attribution for looped transformerscited by 5
  21. 2026Internalizing LLM Reasoning via Discovery and Replay of Latent Actionscited by 4
  22. 2026Latent Thoughts Tuning: Bridging Context and Reasoning with Fused Information in Latent Tokenscited by 4
  23. 2026Why Do LLMs Struggle in Strategic Play? Broken Links Between Observations, Beliefs, and Actionscited by 3
  24. 2026Mem-W: Latent Memory-Native GUI Agentscited by 3
  25. 2026World Models: A Comprehensive Survey of Architectures, Methodologies, Reasoning Paradigms, and Applicationscited by 3
  26. 2026Unlocking the Black Box of Latent Reasoning: An Interpretability-Guided Approach to Interventioncited by 3
  27. 2026What Makes Effective Supervision in Latent Chain-of-Thought: An Information-Theoretic Analysiscited by 3
  28. 2026CoLT: Teaching Multi-Modal Models to Think with Chain of Latent Thoughtscited by 3
  29. 2026Latent Visual Cache for Video Reasoningcited by 3
  30. 2026ThinkTank-ME: A Multi-Expert Framework for Middle East Event Forecastingcited by 2
  31. 2026LatentRouter: Can We Choose the Right Multimodal Model Before Seeing Its Answer?cited by 2
  32. 2026Self-Reference in Large Language Models: The Introspection Threshold for Recursive Self-Improvementcited by 2
  33. 2026Opening the Black Box: A Survey on the Mechanisms of Multi-Step Reasoning in Large Language Modelscited by 1
  34. 2026Latent Abstraction for Retrieval-Augmented Generationcited by 1
  35. 2026CopT: Contrastive On-Policy Thinking with Continuous Spaces for General and Agentic Reasoningcited by 1
  36. 2026Robust and Efficient Guardrails with Latent Reasoningcited by 1
  37. 2026Tyler: Typed Latent Reasoning for Language Models -- When to Think, What to Compute, and How Much to Allocatecited by 1
  38. 2026Formalizing Latent Thoughts: Four Axioms of Thought Representation in LLMscited by 1

50 most cited of 78 All 78 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025A Survey on Latent Reasoning
  2. 2026Latent Reasoning with Normalizing Flows
  3. 2023Igniting Language Intelligence: The Hitchhiker's Guide From Chain-of-Thought Reasoning to Language Agents
  4. 2026Hierarchical Chain-of-Thought: Enhancing LLM Reasoning Performance and Efficiency
  5. 2024Fine-Tuning on Diverse Reasoning Chains Drives Within-Inference CoT Refinement in LLMs
  6. 2024Training Large Language Models to Reason in a Continuous Latent Space
  7. 2026CoLT: Reasoning with Chain of Latent Tool Calls
  8. 2024A Hopfieldian View-based Interpretation for Chain-of-Thought Reasoning