mlsanityData to 7 Oct 2026

Reasoning Aware Self-Consistency: Leveraging Reasoning Paths for Efficient LLM Sampling

arXiv 2408.17017 ↗First posted 30 Aug 2024cs.CL cs.AIcited by 63

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

Abstract (arXiv)

Self-Consistency mitigates hallucinations in Large Language Models (LLMs) by sampling multiple reasoning paths,but it lacks a systematic approach to determine the optimal number of samples or select the most faithful rationale. To address this limitation, we introduce Reasoning-Aware Self-Consistency (RASC), a novel framework that enhances sampling efficiency and reasoning faithfulness by dynamically evaluating both outputs and rationales. RASC assesses the quality of reasoning and the consistency of answers for each generated sample, using these assessments to guide early stopping decisions and rationale selection. The framework employs criteria-based stopping and weighted majority voting, enabling more informed choices on when to halt sampling and which rationale to select. Our comprehensive experiments across diverse question-answering datasets demonstrate that RASC outperforms existing methods, reducing sample usage by approximately 70% while maintaining accuracy. Moreover, RASC facilitates the selection of high-fidelity rationales, thereby improving the faithfulness of LLM outputs. Our approach effectively addresses the efficiency-accuracy trade-off in LLM reasoning tasks, offering a new perspective for more nuanced, faithful, and effective utilization of LLMs in resource-constrained environments.

Key facts. Cited by 63 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 23 of its 38 references that are in the map. First posted on arXiv 30 Aug 2024.

Builds on

23 of 38 references.

Closest ancestors · 5

Same direction, newest first

  1. 2024Integrate the Essence and Eliminate the Dross: Fine-Grained Self-Consistency for Free-Form Language Generation2 Jul
  2. 2024Escape Sky-high Cost: Early-stopping Self-Consistency for Multi-step Reasoning19 Jan
  3. 2023Large Language Models Cannot Self-Correct Reasoning Yet3 Oct
  4. 2023SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoning1 Aug
  5. 2023Let's Sample Step by Step: Adaptive-Consistency for Efficient Reasoning and Coding with LLMs19 May

Other references · 18

  1. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  2. 2022Large Language Models are Zero-Shot Reasonerscited by 4,323
  3. 2023Let's Verify Step by Stepcited by 3,598
  4. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,970
  5. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,881
  6. 2020BLEURT: Learning Robust Metrics for Text Generationcited by 1,196
  7. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 736
  8. 2021BARTScore: Evaluating Generated Text as Text Generationcited by 719
  9. 2024ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Searchcited by 338
  10. 2023How Language Model Hallucinations Can Snowballcited by 310
  11. 2022Large Language Models are Better Reasoners with Self-Verificationcited by 304
  12. 2024Debating with More Persuasive LLMs Leads to More Truthful Answerscited by 255
6 more
  1. 2022ROSCOE: A Suite of Metrics for Scoring Step-by-Step Reasoningcited by 175
  2. 2024The Impact of Reasoning Step Length on Large Language Modelscited by 152
  3. 2023Question Decomposition Improves the Faithfulness of Model-Generated Reasoningcited by 107
  4. 2023Human Feedback is not Gold Standardcited by 82
  5. 2023Can ChatGPT Replace Traditional KBQA Models? An In-depth Analysis of the Question Answering Performance of the GPT LLM Familycited by 71
  6. 2021Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generationcited by 60

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 34; 2026: 27.

63 papers.

Sort:

By area:

  1. 2025Stop Overthinking: A Survey on Efficient Reasoning for Large Language Modelscited by 456
  2. 2025Deep Think with Confidencecited by 199
  3. 2025Efficient Reasoning Models: A Surveycited by 93
  4. 2025Efficient Test-Time Scaling via Self-Calibrationcited by 74
  5. 2025From AI for Science to Agentic Science: A Survey on Autonomous Scientific Discoverycited by 74
  6. 2024Efficiently Scaling LLM Reasoning with Certaindexcited by 56
  7. 2025Budget-Aware Tool Use Enables Effective Agent Scalingcited by 42
  8. 2025Harnessing the Reasoning Economy: A Survey of Efficient Reasoning for Large Language Modelscited by 35
  9. 2025Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMscited by 31
  10. 2025Video-RTS: Rethinking Reinforcement Learning and Test-Time Scaling for Efficient and Enhanced Video Reasoningcited by 29
  11. 2025AlphaOne: Reasoning Models Thinking Slow and Fast at Test Timecited by 27
  12. 2024From Medprompt to o1: Exploration of Run-Time Strategies for Medical Challenge Problems and Beyondcited by 24
38 more
  1. 2025Verifier-free Test-Time Sampling for Vision-Language-Action Modelscited by 21
  2. 2025Lost at the Beginning of Reasoningcited by 15
  3. 2025Prompt-Reverse Inconsistency: LLM Self-Inconsistency Beyond Generative Randomness and Prompt Paraphrasingcited by 14
  4. 2025Generative AI Act II: Test Time Scaling Drives Cognition Engineeringcited by 14
  5. 2025Do Repetitions Matter? Strengthening Reliability in LLM Evaluationscited by 14
  6. 2026Parallel-Probe: Towards Efficient Parallel Thinking via 2D Probingcited by 13
  7. 2025Geometric Uncertainty for Detecting and Correcting Hallucinations in LLMscited by 12
  8. 2025Slim-SC: Thought Pruning for Efficient Scaling with Self-Consistencycited by 12
  9. 2025Certified Self-Consistency: Statistical Guarantees and Test-Time Training for Reliable Reasoning in LLMscited by 12
  10. 2025Exploring the System 1 Thinking Capability of Large Reasoning Modelscited by 10
  11. 2025PiCSAR: Probabilistic Confidence Selection And Ranking for Reasoning Chainscited by 9
  12. 2025Disparities in LLM Reasoning Accuracy and Explanations: A Case Study on African American Englishcited by 8
  13. 2025Autonomous Control Leveraging LLMs: An Agentic Framework for Next-Generation Industrial Automationcited by 8
  14. 2025What makes Reasoning Models Different? Follow the Reasoning Leader for Efficient Decodingcited by 7
  15. 2026Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Modelscited by 7
  16. 2026Optimal Bayesian Stopping for Efficient Inference of Consistent LLM Answerscited by 7
  17. 2026Agentic Test-Time Scaling for WebAgentscited by 7
  18. 2026LLMs Improving LLMs: Agentic Discovery for Test-Time Scalingcited by 7
  19. 2026Re2: Unlocking LLM Reasoning via Reinforcement Learning with Re-solvingcited by 6
  20. 2025PRInTS: Reward Modeling for Long-Horizon Information Seekingcited by 5
  21. 2026Geometry-Calibrated Conformal Abstention for Language Modelscited by 5
  22. 2025Enhancing the Outcome Reward-based RL Training of MLLMs with Self-Consistency Samplingcited by 4
  23. 2025Beg to Differ: Understanding Reasoning-Answer Misalignment Across Languagescited by 4
  24. 2026TapSampling: Inference-Time Sampling with a Task-Progress-Understanding Verifier for Robotic Manipulationcited by 4
  25. 2025OptScale: Probabilistic Optimality for Inference-time Scalingcited by 3
  26. 2025ATTS: Asynchronous Test-Time Scaling via Conformal Predictioncited by 3
  27. 2025Evaluating Large Language Models for Detecting Antisemitismcited by 3
  28. 2025Explore-Execute Chain: Towards an Efficient Structured Reasoning Paradigmcited by 2
  29. 2026Finding the Cracks: Improving LLMs Reasoning with Paraphrastic Probing and Consistency Verificationcited by 2
  30. 2026PETS: A Principled Framework Towards Optimal Trajectory Allocation for Efficient Test-Time Self-Consistencycited by 2
  31. 2026Beyond Majority Voting: Efficient Best-Of-N with Radial Consensus Scorecited by 2
  32. 2025FedCoT: Communication-Efficient Federated Reasoning Enhancement for Large Language Modelscited by 1
  33. 2026The Geometric Reasoner: Manifold-Informed Latent Foresight Search for Long-Context Reasoningcited by 1
  34. 2026Benchmarking Large Language Models for Knowledge Graph Validationcited by 1
  35. 2026RoboEvolve: Co-Evolving Planner-Simulator for Robotic Manipulation with Limited Datacited by 1
  36. 2026When Does Delegation Beat Majority? A Delegation-Based Aggregator for Multi-Sample LLM Inferencecited by 1
  37. 2026More Correct Mass, Worse Answers: Why Power Sampling Can Fail and How to Fix Itcited by 1
  38. 2026ParaTempo: Efficient Parallel Reasoning via Temporal Confidencecited by 1

50 most cited of 63 All 63 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Optimal Self-Consistency for Efficient Reasoning with Large Language Models
  2. 2025A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning
  3. 2024Lachesis: Predicting LLM Inference Accuracy using Structural Properties of Reasoning Paths
  4. 2024Semantic Self-Consistency: Enhancing Language Model Reasoning via Semantic Weighting
  5. 2026Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning
  6. 2026Self-Consistency from Only Two Samples: CoT-PoT Ensembling for Efficient LLM Reasoning
  7. 2026Adaptive Power Sampling for LLM Reasoning
  8. 2026Enhancing reasoning accuracy in large language models during inference time