mlsanityData to 14 Oct 2026

Temporal Consistency for LLM Reasoning Process Error Identification

arXiv 2503.14495 ↗First posted 18 Mar 2025cs.CL cs.AI cs.LGcited by 10

On the map: LLM reasoning → Mathematical reasoning

Abstract (arXiv)

Verification is crucial for effective mathematical reasoning. We present a new temporal consistency method where verifiers iteratively refine their judgments based on the previous assessment. Unlike one-round verification or multi-model debate approaches, our method leverages consistency in a sequence of self-reflection actions to improve verification accuracy. Empirical evaluations across diverse mathematical process error identification benchmarks (Mathcheck, ProcessBench, and PRM800K) show consistent performance improvements over baseline methods. When applied to the recent DeepSeek R1 distilled models, our method demonstrates strong performance, enabling 7B/8B distilled models to outperform all 70B/72B models and GPT-4o on ProcessBench. Notably, the distilled 14B model with our method achieves performance comparable to Deepseek-R1. Our codes are available at https://github.com/jcguo123/Temporal-Consistency

Key facts. Cited by 10 papers in the map (data through 14 Oct 2026). Placed in Mathematical reasoning, one of 17 directions in LLM reasoning. Builds on 41 of its 51 references that are in the map. First posted on arXiv 18 Mar 2025.

Builds on

41 of 51 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking8 Jan
  2. 2024Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models10 Oct
  3. 2024Is Your Model Really A Good Math Reasoner? Evaluating Mathematical Reasoning with Checklist11 Jul
  4. 2024MindStar: Enhancing Math Reasoning in Pre-trained LLMs at Inference Time25 May

Other references · 37

  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. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  4. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  5. 2024Qwen2.5 Technical Reportcited by 4,942
  6. 2024GPT-4o System Cardcited by 4,588
  7. 2023Let's Verify Step by Stepcited by 3,789
  8. 2023Mistral 7Bcited by 3,164
  9. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  10. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  11. 2025s1: Simple test-time scalingcited by 1,341
  12. 2024OpenAI o1 System Cardcited by 1,244
25 more
  1. 2024OlympiadBench: A Challenging Benchmark for Promoting AGI with Olympiad-Level Bilingual Multimodal Scientific Problemscited by 1,136
  2. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,090
  3. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 1,017
  4. 2022Automatic Chain of Thought Prompting in Large Language Modelscited by 688
  5. 2024Generative Verifiers: Reward Modeling as Next-Token Predictioncited by 480
  6. 2024Improve Mathematical Reasoning in Language Models by Automated Process Supervisioncited by 441
  7. 2024ProcessBench: Identifying Process Errors in Mathematical Reasoningcited by 292
  8. 2024Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Modelscited by 290
  9. 2024Rewarding Progress: Scaling Automated Process Verifiers for LLM Reasoningcited by 279
  10. 2023SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoningcited by 159
  11. 2024Stream of Search (SoS): Learning to Search in Languagecited by 145
  12. 2025Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scalingcited by 136
  13. 2024ARGS: Alignment as Reward-Guided Searchcited by 103
  14. 2025Multiagent Finetuning: Self Improvement with Diverse Reasoning Chainscited by 95
  15. 2025T1: Advancing Language Model Reasoning through Reinforcement Learning and Inference Scalingcited by 77
  16. 2025Unlocking Multimodal Mathematical Reasoning via Process Reward Modelcited by 65
  17. 2025ReasonFlux: Hierarchical LLM Reasoning via Scaling Thought Templatescited by 63
  18. 2024TreeBoN: Enhancing Inference-Time Alignment with Speculative Tree-Search and Best-of-N Samplingcited by 61
  19. 2025Evolving Deeper LLM Thinkingcited by 58
  20. 2024SuperCorrect: Advancing Small LLM Reasoning with Thought Template Distillation and Self-Correctioncited by 40
  21. 2025VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Datacited by 40
  22. 2024Critic-CoT: Boosting the reasoning abilities of large language model via Chain-of-thoughts Criticcited by 28
  23. 2024On the Limited Generalization Capability of the Implicit Reward Model Induced by Direct Preference Optimizationcited by 27
  24. 2024Mirror-Consistency: Harnessing Inconsistency in Majority Votingcited by 17
  25. 2025Self-Evolving Critique Abilities in Large Language Modelscited by 12

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 5.

10 papers.

  1. 2025Thought calibration: Efficient and confident test-time scalingcited by 21
  2. 2025Stepwise Guided Policy Optimization: Coloring your Incorrect Reasoning in GRPOcited by 19
  3. 2025Reliable Fine-Grained Evaluation of Natural Language Math Proofscited by 18
  4. 2025Toward Scientific Reasoning in LLMs: Training from Expert Discussions via Reinforcement Learningcited by 7
  5. 2026More Rounds, More Noise: Why Multi-Turn Review Fails to Improve Cross-Context Verificationcited by 4
  6. 2026RAG or Learning? Understanding the Limits of LLM Adaptation under Continuous Knowledge Drift in the Real Worldcited by 4
  7. 2025On the Role of Preference Variance in Preference Optimizationcited by 3
  8. 2026"I May Not Have Articulated Myself Clearly": Diagnosing Dynamic Instability in LLM Reasoning at Inference Timecited by 1
  9. 2026Probing to Refine: Reinforcement Distillation of LLMs via Explanatory Inversioncited by 1
  10. 2026MMTClinic: Multimodal, Multilingual Time Series Question Answering and Reasoning Benchmark for Clinical Domaincited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Beyond the Answer Key: Robustness Evaluation of Large Language Models for Step-Level Mathematical Verification
  2. 2025Safe: Enhancing Mathematical Reasoning in Large Language Models via Retrospective Step-aware Formal Verification
  3. 2025MathBode: Measuring the Stability of LLM Reasoning using Frequency Response
  4. 2024Evaluating Mathematical Reasoning of Large Language Models: A Focus on Error Identification and Correction
  5. 2024Improving LLM Reasoning through Scaling Inference Computation with Collaborative Verification
  6. 2026Improving Mathematical Reasoning Capabilities in Large Language Models via Reasoning Process Error Classification
  7. 2025ReliableMath: Benchmark of Reliable Mathematical Reasoning on Large Language Models
  8. 2026Pipeline for Verifying LLM-Generated Mathematical Solutions