mlsanityData to 7 Oct 2026

Contrastive Chain-of-Thought Prompting

arXiv 2311.09277 ↗First posted 15 Nov 2023cs.CLcited by 52

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

Abstract (arXiv)

Despite the success of chain of thought in enhancing language model reasoning, the underlying process remains less well understood. Although logically sound reasoning appears inherently crucial for chain of thought, prior studies surprisingly reveal minimal impact when using invalid demonstrations instead. Furthermore, the conventional chain of thought does not inform language models on what mistakes to avoid, which potentially leads to more errors. Hence, inspired by how humans can learn from both positive and negative examples, we propose contrastive chain of thought to enhance language model reasoning. Compared to the conventional chain of thought, our approach provides both valid and invalid reasoning demonstrations, to guide the model to reason step-by-step while reducing reasoning mistakes. To improve generalization, we introduce an automatic method to construct contrastive demonstrations. Our experiments on reasoning benchmarks demonstrate that contrastive chain of thought can serve as a general enhancement of chain-of-thought prompting.

Key facts. Cited by 52 papers in the map (data through 7 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 5 of its 26 references that are in the map. First posted on arXiv 15 Nov 2023.

Builds on

5 of 26 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Navigate through Enigmatic Labyrinth A Survey of Chain of Thought Reasoning: Advances, Frontiers and Future27 Sep

Other references · 4

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2020Scaling Laws for Neural Language Modelscited by 6,411
  3. 2021Scaling Language Models: Methods, Analysis & Insights from Training Gophercited by 1,199
  4. 2021It Takes Two to Tango: Combining Visual and Textual Information for Detecting Duplicate Video-Based Bug Reportscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 23; 2025: 20; 2026: 8.

52 papers.

Sort:

By area:

  1. 2024A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applicationscited by 428
  2. 2025Reasoning Models Don't Always Say What They Thinkcited by 388
  3. 2023Efficient Large Language Models: A Surveycited by 183
  4. 2024Recursive Introspection: Teaching Language Model Agents How to Self-Improvecited by 162
  5. 2025Are DeepSeek R1 And Other Reasoning Models More Faithful?cited by 69
  6. 2024Key-Point-Driven Data Synthesis with its Enhancement on Mathematical Reasoningcited by 58
  7. 2024A Survey of Prompt Engineering Methods in Large Language Models for Different NLP Taskscited by 46
  8. 2024A Theoretical Understanding of Self-Correction through In-context Alignmentcited by 43
  9. 2024In-Context Principle Learning from Mistakescited by 42
  10. 2024Can Language Models Perform Robust Reasoning in Chain-of-thought Prompting with Noisy Rationales?cited by 40
  11. 2024Unleashing LLM Reasoning Capability via Scalable Question Synthesis from Scratchcited by 24
  12. 2025From Debate to Equilibrium: Belief-Driven Multi-Agent LLM Reasoning via Bayesian Nash Equilibriumcited by 24
38 more
  1. 2025Activation Steering for Chain-of-Thought Compressioncited by 22
  2. 2024Distillation Contrastive Decoding: Improving LLMs Reasoning with Contrastive Decoding and Distillationcited by 16
  3. 2024Enhancing Chain of Thought Prompting in Large Language Models via Reasoning Patternscited by 13
  4. 2024Focus on Your Question! Interpreting and Mitigating Toxic CoT Problems in Commonsense Reasoningcited by 11
  5. 2024Large Language Models are Contrastive Reasonerscited by 11
  6. 2024Rethinking Chain-of-Thought from the Perspective of Self-Trainingcited by 10
  7. 2025Truthful or Fabricated? Using Causal Attribution to Mitigate Reward Hacking in Explanationscited by 9
  8. 2025Toward Structured Knowledge Reasoning: Contrastive Retrieval-Augmented Generation on Experiencecited by 7
  9. 2024Baichuan Alignment Technical Reportcited by 6
  10. 2024Chain-of-Thought in Large Language Models: Decoding, Projection, and Activationcited by 6
  11. 2024Computational Argumentation-based Chatbots: a Surveycited by 5
  12. 2025ReFineVLA: Reasoning-Aware Teacher-Guided Transfer Fine-Tuningcited by 5
  13. 2024"Let's Argue Both Sides": Argument Generation Can Force Small Models to Utilize Previously Inaccessible Reasoning Capabilitiescited by 4
  14. 2025Leveraging the Power of Large Language Models in Entity Linking via Adaptive Routing and Targeted Reasoningcited by 4
  15. 2026Revisiting Chain-of-Thought Reasoning under Limited Supervision: Semi-supervised Chain-of-Thought Learningcited by 4
  16. 2024Can formal argumentative reasoning enhance LLMs performances?cited by 3
  17. 2024RAVEL: Rare Concept Generation and Editing via Graph-driven Relational Guidancecited by 3
  18. 2025Improving RAG for Personalization with Author Features and Contrastive Examplescited by 3
  19. 2025FOL-Traces: Verified First-Order Logic Reasoning Traces at Scalecited by 3
  20. 2026ReasonAlloc: Hierarchical Decoding-Time KV Cache Budget Allocation for Reasoning Modelscited by 3
  21. 2025FLAMES: Improving LLM Math Reasoning via a Fine-Grained Analysis of the Data Synthesis Pipelinecited by 2
  22. 2025Mitigating Strategy-Selection Bias in Reasoning for More Effective Test-Time Scalingcited by 2
  23. 2024Logic Contrastive Reasoning with Lightweight Large Language Model for Math Word Problemscited by 1
  24. 2024AutoReason: Automatic Few-Shot Reasoning Decompositioncited by 1
  25. 2025Zero-Shot Image-Based Large Language Model Approach to Road Pavement Monitoringcited by 1
  26. 2025The Challenge of Teaching Reasoning to LLMs Without RL or Distillationcited by 1
  27. 2025How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activationcited by 1
  28. 2025Multi-Agent Multimodal Large Language Model Framework for Automated Interpretation of Fuel Efficiency Analytics in Public Transportationcited by 1
  29. 2026ReFineVLA: Multimodal Reasoning-Aware Generalist Robotic Policies via Teacher-Guided Fine-Tuningcited by 1
  30. 2026Chain-of-Experience for Continual LLM Improvementcited by 1
  31. 2024FRoG: Evaluating Fuzzy Reasoning of Generalized Quantifiers in Large Language Modelscited by 0
  32. 2024An Overview and Discussion on Using Large Language Models for Implementation Generation of Solutions to Open-Ended Problemscited by 0
  33. 2025Learning Virtual Machine Scheduling in Cloud Computing through Language Agentscited by 0
  34. 2025On the Thinking-Language Modeling Gap in Large Language Modelscited by 0
  35. 2025Automatic Posology Structuration : What role for LLMs?cited by 0
  36. 2025Engineering Sentiencecited by 0
  37. 2026CAP-CoT: Cycle Adversarial Prompt for Improving Chain of Thoughts in LLM Reasoningcited by 0
  38. 2026Synthetic Contrastive Reasoning for Multi-Table Q&Acited by 0

50 most cited of 52 All 52 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  2. 2024Chain-of-Thought Augmentation with Logit Contrast for Enhanced Reasoning in Language Models
  3. 2022Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters
  4. 2023Boosting Language Models Reasoning with Chain-of-Knowledge Prompting
  5. 2023Why Can Large Language Models Generate Correct Chain-of-Thoughts?
  6. 2026Does Verbose Chain-of-Thought Really Help? In-Distribution Evidence that Content, Not Length, Matters
  7. 2024A Theoretical Understanding of Chain-of-Thought: Coherent Reasoning and Error-Aware Demonstration
  8. 2023Symbolic Chain-of-Thought Distillation: Small Models Can Also "Think" Step-by-Step