mlsanityData to 7 Oct 2026

SSR: Socratic Self-Refine for Large Language Model Reasoning

arXiv 2511.10621 ↗First posted 13 Nov 2025cs.CL cs.AI cs.LGcited by 6

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Large Language Models (LLMs) have demonstrated remarkable reasoning abilities, yet existing test-time frameworks often rely on coarse self-verification and self-correction, limiting their effectiveness on complex tasks. In this paper, we propose Socratic Self-Refine (SSR), a novel framework for fine-grained evaluation and precise refinement of LLM reasoning. Our proposed SSR decomposes model responses into verifiable (sub-question, sub-answer) pairs, enabling step-level confidence estimation through controlled re-solving and self-consistency checks. By pinpointing unreliable steps and iteratively refining them, SSR produces more accurate and interpretable reasoning chains. Empirical results across five reasoning benchmarks and three LLMs show that SSR consistently outperforms state-of-the-art iterative self-refinement baselines. Beyond performance gains, SSR provides a principled black-box approach for evaluating and understanding the internal reasoning processes of LLMs. Code is available at https://github.com/SalesforceAIResearch/socratic-self-refine-reasoning.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 38 of its 57 references that are in the map. First posted on arXiv 13 Nov 2025.

Builds on

38 of 57 references.

Closest ancestors · 11

Same direction, newest first

  1. 2025Deep Think with Confidence21 Aug
  2. 2025Probabilistic Soundness Guarantees in LLM Reasoning Chains17 Jul
  3. 2025Solve-Detect-Verify: Inference-Time Scaling with Flexible Generative Verifier17 May
  4. 2025TokUR: Token-Level Uncertainty Estimation for Large Language Model Reasoning16 May
  5. 2025Efficient Test-Time Scaling via Self-Calibration25 Feb
  6. 2025Scalable Best-of-N Selection for Large Language Models via Self-Certainty25 Feb
  7. 2025Atom of Thoughts for Markov LLM Test-Time Scaling17 Feb
  8. 2025SETS: Leveraging Self-Verification and Self-Correction for Improved Test-Time Scaling31 Jan
  9. 2024Forest-of-Thought: Scaling Test-Time Compute for Enhancing LLM Reasoning12 Dec
  10. 2024Language Models are Hidden Reasoners: Unlocking Latent Reasoning Capabilities via Self-Rewarding6 Nov
  11. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug

Other references · 27

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  4. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  5. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  6. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  7. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  8. 2022Emergent Abilities of Large Language Modelscited by 2,483
  9. 2024A Survey on LLM-as-a-Judgecited by 1,384
  10. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 787
  11. 2025Humanity's Last Examcited by 467
  12. 2025The Lessons of Developing Process Reward Models in Mathematical Reasoningcited by 402
15 more
  1. 2024Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judgecited by 249
  2. 2025When More is Less: Understanding Chain-of-Thought Length in LLMscited by 182
  3. 2024Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8Bcited by 156
  4. 2024Mutual Reasoning Makes Smaller LLMs Stronger Problem-Solverscited by 137
  5. 2025REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewardscited by 121
  6. 2025A Survey of Frontiers in LLM Reasoning: Inference Scaling, Learning to Reason, and Agentic Systemscited by 108
  7. 2024Physics of Language Models: Part 2.1, Grade-School Math and the Hidden Reasoning Processcited by 107
  8. 2025GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoningcited by 94
  9. 2024Explaining Length Bias in LLM-Based Preference Evaluationscited by 48
  10. 2025Evaluating Judges as Evaluators: The JETTS Benchmark of LLM-as-Judges as Test-Time Scaling Evaluatorscited by 42
  11. 2025Premise-Augmented Reasoning Chains Improve Error Identification in Math reasoning with LLMscited by 22
  12. 2025Uncertainty-Aware Step-wise Verification with Generative Reward Modelscited by 21
  13. 2025Variation in Verification: Understanding Verification Dynamics in Large Language Modelscited by 18
  14. 2025Graph of Verification: Structured Verification of LLM Reasoning with Directed Acyclic Graphscited by 8
  15. 2025Socratic-MCTS: Test-Time Visual Reasoning by Asking the Right Questionscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 6.

6 papers.

  1. 2026Code as Agent Harnesscited by 40
  2. 2026Revision or Re-Solving? Decomposing Second-Pass Gains in Multi-LLM Pipelinescited by 7
  3. 2026More Rounds, More Noise: Why Multi-Turn Review Fails to Improve Cross-Context Verificationcited by 4
  4. 2026Spectral Origins of the Self-Correction Blind Spot in Autoregressive Generationcited by 0
  5. 2026Recursive Agentic Reasoningcited by 0
  6. 2026Evidence-Inference Reconstruction: When The Evidence Is Recalled But The Reasoning Goes Wrongcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025RefineBench: Evaluating Refinement Capability of Language Models via Checklists
  2. 2024Small Language Models Need Strong Verifiers to Self-Correct Reasoning
  3. 2025Self-Reasoning Language Models: Unfold Hidden Reasoning Chains with Few Reasoning Catalyst
  4. 2026Enhancing reasoning accuracy in large language models during inference time
  5. 2024S^3cMath: Spontaneous Step-level Self-correction Makes Large Language Models Better Mathematical Reasoners
  6. 2024MC-NEST: Enhancing Mathematical Reasoning in Large Language Models leveraging a Monte Carlo Self-Refine Tree
  7. 2025Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
  8. 2025Self-Reflective Generation at Test Time