mlsanityData to 14 Oct 2026

Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketching

arXiv 2503.05179 ↗First posted 7 Mar 2025cs.CL cs.AI cs.LGcited by 117

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

Abstract (arXiv)

Recent advances in large language models (LLMs) have enabled strong reasoning capabilities through Chain-of-Thought (CoT) prompting, which elicits step-by-step problem solving, but often at the cost of excessive verbosity in intermediate outputs, leading to increased computational overhead. We propose Sketch-of-Thought (SoT), a prompting framework that integrates cognitively inspired reasoning paradigms with linguistic constraints to reduce token usage while preserving reasoning accuracy. SoT is designed as a flexible, modular approach and is instantiated with three paradigms--Conceptual Chaining, Chunked Symbolism, and Expert Lexicons--each tailored to distinct reasoning tasks and selected dynamically at test-time by a lightweight routing model. Across 18 reasoning datasets spanning multiple domains, languages, and modalities, SoT achieves token reductions of up to 84% with minimal accuracy loss. In tasks such as mathematical and multi-hop reasoning, it even improves accuracy while shortening outputs.

Key facts. Cited by 117 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 38 of its 56 references that are in the map. First posted on arXiv 7 Mar 2025.

Builds on

38 of 56 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Chain of Draft: Thinking Faster by Writing Less25 Feb
  2. 2024To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoning18 Sep
  3. 2023Fewer is More: Boosting LLM Reasoning with Reinforced Context Pruning14 Dec
  4. 2023SCOTT: Self-Consistent Chain-of-Thought Distillation3 May
  5. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Models21 Mar

Other references · 33

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  4. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  5. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,307
  6. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  7. 2023FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioningcited by 2,598
  8. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,592
  9. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
  10. 2018Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answeringcited by 2,164
  11. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,919
  12. 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,822
21 more
  1. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  2. 2020What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Examscited by 1,356
  3. 2019PubMedQA: A Dataset for Biomedical Research Question Answeringcited by 1,288
  4. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 916
  5. 2019DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphscited by 903
  6. 2021Are NLP Models really able to Solve Simple Math Word Problems?cited by 878
  7. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 743
  8. 2023Automatic Prompt Optimization with "Gradient Descent" and Beam Searchcited by 653
  9. 2017Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problemscited by 650
  10. 2019QASC: A Dataset for Question Answering via Sentence Compositioncited by 252
  11. 2023Specializing Smaller Language Models towards Multi-Step Reasoningcited by 251
  12. 2025Training Language Models to Reason Efficientlycited by 247
  13. 2020ReClor: A Reading Comprehension Dataset Requiring Logical Reasoningcited by 217
  14. 2020Retrieve, Program, Repeat: Complex Knowledge Base Question Answering via Alternate Meta-learningcited by 216
  15. 2023Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoningcited by 143
  16. 2024Concise Thoughts: Impact of Output Length on LLM Reasoning and Costcited by 109
  17. 2023A Survey of Reasoning with Foundation Modelscited by 70
  18. 2018WorldTree: A Corpus of Explanation Graphs for Elementary Science Questions supporting Multi-Hop Inferencecited by 60
  19. 2023LeanContext: Cost-Efficient Domain-Specific Question Answering Using LLMscited by 40
  20. 2024Self-Refine Instruction-Tuning for Aligning Reasoning in Language Modelscited by 24
  21. 2024Do Language Models Exhibit Human-like Structural Priming Effects?cited by 21

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 77; 2026: 40.

117 papers.

Sort:

By area:

  1. 2025Stop Overthinking: A Survey on Efficient Reasoning for Large Language Modelscited by 464
  2. 2025A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?cited by 223
  3. 2025Dynamic Early Exit in Reasoning Modelscited by 219
  4. 2025Reasoning Models Can Be Effective Without Thinkingcited by 187
  5. 2025Multimodal Chain-of-Thought Reasoning: A Comprehensive Surveycited by 182
  6. 2025A Survey of Efficient Reasoning for Large Reasoning Models: Language, Multimodality, and Beyondcited by 132
  7. 2025Thinkless: LLM Learns When to Thinkcited by 113
  8. 2025Efficient Reasoning Models: A Surveycited by 95
  9. 2025S-GRPO: Early Exit via Reinforcement Learning in Reasoning Modelscited by 87
  10. 2025AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learningcited by 87
  11. 2025Think Silently, Think Fast: Dynamic Latent Compression of LLM Reasoning Chainscited by 85
  12. 2025ShorterBetter: Guiding Reasoning Models to Find Optimal Inference Length for Efficient Reasoningcited by 72
38 more
  1. 2025Efficient Inference for Large Reasoning Models: A Surveycited by 61
  2. 2025Speculative Thinking: Enhancing Small-Model Reasoning with Large Model Guidance at Inference Timecited by 48
  3. 2025SelfBudgeter: Adaptive Token Allocation for Efficient LLM Reasoningcited by 48
  4. 2025Answer Convergence as a Signal for Early Stopping in Reasoningcited by 40
  5. 2025Long-Short Chain-of-Thought Mixture Supervised Fine-Tuning Eliciting Efficient Reasoning in Large Language Modelscited by 39
  6. 2025Optimizing Length Compression in Large Reasoning Modelscited by 38
  7. 2025Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Modelscited by 37
  8. 2025Cost-of-Pass: An Economic Framework for Evaluating Language Modelscited by 35
  9. 2025ConCISE: Confidence-guided Compression in Step-by-step Efficient Reasoningcited by 33
  10. 2025Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMscited by 31
  11. 2025FlashThink: An Early Exit Method For Efficient Reasoningcited by 28
  12. 2025Let LRMs Break Free from Overthinking via Self-Braking Tuningcited by 27
  13. 2025When to Continue Thinking: Adaptive Thinking Mode Switching for Efficient Reasoningcited by 26
  14. 2025Shorten After You're Right: Lazy Length Penalties for Reasoning RLcited by 25
  15. 2025Route to Reason: Adaptive Routing for LLM and Reasoning Strategy Selectioncited by 23
  16. 2025Efficient Reasoning for LLMs through Speculative Chain-of-Thoughtcited by 22
  17. 2025ThinKV: Thought-Adaptive KV Cache Compression for Efficient Reasoning Modelscited by 21
  18. 2025CP-Router: An Uncertainty-Aware Router Between LLM and LRMcited by 20
  19. 2025A Survey of Slow Thinking-based Reasoning LLMs using Reinforced Learning and Inference-time Scaling Lawcited by 19
  20. 2026Does Your Reasoning Model Implicitly Know When to Stop Thinking?cited by 19
  21. 2025The Price of a Second Thought: On the Evaluation of Reasoning Efficiency in Large Language Modelscited by 18
  22. 2025Towards Concise and Adaptive Thinking in Large Reasoning Models: A Surveycited by 18
  23. 2025ThinkLess: A Training-Free Inference-Efficient Method for Reducing Reasoning Redundancycited by 17
  24. 2025Not All Thoughts are Generated Equal: Efficient LLM Reasoning via Multi-Turn Reinforcement Learningcited by 16
  25. 2025Two Experts Are All You Need for Steering Thinking: Reinforcing Cognitive Effort in MoE Reasoning Models Without Additional Trainingcited by 16
  26. 2025LazyEviction: Lagged KV Eviction with Attention Pattern Observation for Efficient Long Reasoningcited by 16
  27. 2025DRPO: Efficient Reasoning via Decoupled Reward Policy Optimizationcited by 16
  28. 2025Generative AI Act II: Test Time Scaling Drives Cognition Engineeringcited by 15
  29. 2025Concise Reasoning, Big Gains: Pruning Long Reasoning Trace with Difficulty-Aware Promptingcited by 15
  30. 2025Thinking with Visual Abstract: Enhancing Multimodal Reasoning via Visual Abstractioncited by 15
  31. 2025MALLM: Multi-Agent Large Language Models Frameworkcited by 15
  32. 2025From Efficiency to Adaptivity: A Deeper Look at Adaptive Reasoning in Large Language Modelscited by 12
  33. 2025Exploring the System 1 Thinking Capability of Large Reasoning Modelscited by 10
  34. 2025Efficient Reasoning Through Suppression of Self-Affirmation Reflections in Large Reasoning Modelscited by 10
  35. 2025ConciseHint: Boosting Efficient Reasoning via Continuous Concise Hints during Generationcited by 10
  36. 2025Controlling Thinking Speed in Reasoning Modelscited by 10
  37. 2025DeepPrune: Parallel Scaling without Inter-trace Redundancycited by 10
  38. 2025Revisiting Model Interpolation for Efficient Reasoningcited by 10

50 most cited of 117 All 117 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2022Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  3. 2025Mapping the Minds of LLMs: A Graph-Based Analysis of Reasoning LLM
  4. 2025Reasoning Beyond Language: A Comprehensive Survey on Latent Chain-of-Thought Reasoning
  5. 2024Break the Chain: Large Language Models Can be Shortcut Reasoners
  6. 2024Enhancing Chain of Thought Prompting in Large Language Models via Reasoning Patterns
  7. 2024Self-Harmonized Chain of Thought
  8. 2025Focused Chain-of-Thought: Efficient LLM Reasoning via Structured Input Information