Self-Consistency in Language Models

Follow this topic Ask about this topic

This topic studies how language models’ answers and reasoning vary across repeated generations, prompts, and related inputs, and whether consistency can improve reliability. Common approaches sample multiple reasoning paths and aggregate them with majority voting, semantic clustering, ranking, or verification; adaptive sampling and stopping rules aim to reduce the cost. Other work trains models to align their explanations and behavior or tests consistency in settings such as safety, preferences, and evaluation. A central tension is that agreement can improve performance but may also reinforce correlated errors, so consistency alone is not proof of correctness.

Latest papers

49 papers10 papers in the last four weeks, up 67% on the four weeks before. 0.1% of all new papers.

Reliable Chain-of-Thought via Prefix Consistency
May 8, 2026Naoto Iwase, Yuki Ichihara, Mohammad Atif Quamar +1Self-Consistency Decoding Reasoning Consistency in Language Models
Nagoya University · Nara Institute of Science and Technology · Mohamed bin Zayed University of Artificial Intelligence +1
Binary Agreement for Efficient Unsupervised Confidence Calibration of Reasoning LLMs
Apr 21, 2026Thomas Zollo, Jimmy Wang, Richard ZemelConfidence Estimation in Language Models Language Model Calibration
Columbia University
Self-Consistency from Only Two Samples: CoT-PoT Ensembling for Efficient LLM Reasoning
Apr 19, 2026Raman Saparkhan, Majd Hawasly, Md Rizwan Parvez +1LLM Inference Efficiency Self-Consistency Decoding
Carnegie Mellon University Qatar · Qatar Computing Research Institute
Efficient Best-Of-N with Radial Consensus Score
Apr 14, 2026Manh Nguyen, Sunil Gupta, Hung LeBest-of-N Sampling LLM Reranking
Applied Artificial Intelligence Initiative, Deakin University, Australia
Consistency of AI-Generated Exercise Prescriptions: A Repeated Generation Study Using a Large Language Model
Apr 13, 2026Kihyuk LeeSelf-Consistency in Language Models Medical LLMs
Data Convergence Team, Office of Hospital Information, Seoul National University Bundang Hospital, Seongnam, Republic of Korea
Frugal Knowledge Graph Construction with Local LLMs: A Zero-Shot Pipeline, Self-Consistency and Wisdom of Artificial Crowds
Apr 13, 2026Pierre JourlinKG Construction Knowledge Graph Question Answering
LIA · Avignon Université, Laboratoire d’Informatique d’Avignon (LIA)
The Impact of LLM Self-Consistency and Reasoning Effort on Automated Scoring Accuracy and Cost
Apr 3, 2026Scott FrohnMathematics Education Automated Evaluation
Khan Academy, Mountain View, CA, USA
Do We Need Frontier Models to Verify Mathematical Proofs?
Apr 2, 2026Aaditya Naik, Guruprerana Shabadi, Rajeev Alur +1LLM Input Optimization Reasoning Verification
University of Pennsylvania, U.S.A.
Beyond Symbolic Solving: Multi Chain-of-Thought Voting for Geometric Reasoning in Large Language Models
Apr 1, 2026Md. Abu Bakor Siddique, Shahrin Hossain, Sadman Ahmed Siam +3LLM Mathematical Reasoning Self-Consistency in Language Models
Systems and Software Lab (SSL), Department of Computer Science and Engineering, Islamic University of Technology, Dhaka, Bangladesh
How Uncertainty Estimation Scales with Sampling in Reasoning Models
Mar 19, 2026Maksym Del, Markus Kängsepp, Marharyta Domnich +4LLM Uncertainty Estimation Uncertainty Quantification
Institute of Computer Science, University of Tartu
CoVerRL: Breaking the Consensus Trap in Label-Free Reasoning via Generator-Verifier Co-Evolution
Mar 18, 2026Teng Pan, Yuchen Yan, Zixuan Wang +6Language Model Self-Verification RL for Language Model Reasoning
Zhejiang University · Baidu Inc.
Dual Consensus: Escaping from Spurious Majority in Unsupervised RLVR via Two-Stage Vote Mechanism
Mar 17, 2026Kaixuan Du, Meng Cao, Hang Zhang +3RL for Language Model Reasoning Self-Training for Language Models
School of Automation Science and Electrical Engineering, Beihang University
Toward Epistemic Stability: Engineering Consistent Procedures for Industrial LLM Hallucination Reduction
Mar 8, 2026Brian Freeman, Adam Kicklighter, Matt Erdman +1LLM Hallucination Mitigation LLM Prompting
Trane Technologies, Davidson NC, USA
nchellwig at SemEval-2026 Task 3: Self-Consistent Structured Generation (SCSG) for Dimensional Aspect-Based Sentiment Analysis using Large Language Models
Mar 2, 2026Nils Constantin Hellwig, Jakob Fehle, Udo Kruschwitz +1Aspect-Based Sentiment Analysis Structured Output Generation
Media Informatics Group, University of Regensburg, Regensburg, Germany · Information Science Group, University of Regensburg, Regensburg, Germany
Consensus is Not Verification: Why Crowd Wisdom Strategies Fail for LLM Truthfulness
Feb 20, 2026Yegor Denisov-Blanch, Joshua Kazdan, Jessica Chudnovsky +4Correlated Errors in Language Models Inference-Time Scaling
Stanford Computer Science · Stanford Statistics · Stanford University +1
PETS: A Principled Framework Towards Optimal Trajectory Allocation for Efficient Test-Time Self-Consistency
Feb 18, 2026Zhangyi Liu, Huaizhi Qu, Xiaowei Yin +4Inference-Time Compute Allocation Self-Consistency in Language Models
Stanford University · UNC at Chapel Hill · Yale University +1
Optimal Self-Consistency for Efficient Reasoning with Large Language Models
Nov 15, 2025Austin Feng, Marius Alonso, Ambroise Odonnat +2Self-Consistency Decoding Inference-Time Scaling
1Yale University, New Haven, CT, USA · 2Noah’s Ark Lab, Paris, France · 3Inria, Rennes, France +1
Sample, Align, Synthesize: Graph-Based Response Synthesis with ConGrs
Oct 3, 2025Sayan Ghosh, Shahzaib Saqib Warraich, Dhruv Tarsadiya +2Text Generation Self-Consistency in Language Models
Thomas Lord Department of Computer Science, University of Southern California
Atomic Consistency Preference Optimization for Long-Form Question Answering
May 14, 2025Jingfeng Chen, Raghuveer Thirukovalluru, Junlin Wang +2LLM Hallucination Mitigation LLM Alignment

May 8, 2026J/K move · Enter open · S save

Reliable Chain-of-Thought via Prefix Consistency

Naoto Iwase, Yuki Ichihara, Mohammad Atif Quamar, Junpei Komiyama

Nagoya University · Nara Institute of Science and Technology · Mohamed bin Zayed University of Artificial Intelligence+1

Self-Consistency Decoding Reasoning Consistency in Language Models LLM Reliability Efficient Language Model Inference Self-Consistency in Language Models LLM Reasoning+2

Open paper AskPDF arXiv

Large Language Models often improve accuracy on reasoning tasks by sampling multiple Chain-of-Thought (CoT) traces and aggregating them with majority voting (MV), a test-time technique called self-consistency. When we truncate a CoT partway through and regenerate the remainder, we observe that traces with correct answers reproduce their original answer more often than traces with wrong answers. We use this difference as a reliability signal, prefix consistency, that weights each candidate answer by how often it reappears under regeneration. It requires no access to token log-probabilities or self-rating prompts. Across five reasoning models and four math and science benchmarks, prefix consistency is the best correctness predictor in most settings, and reweighting votes by it reaches Standard MV plateau accuracy at up to 21x fewer tokens (median 4.6x). Our code is available at https://github.com/naoto-iwase/prefix-consistency.