mlsanityData to 13 Oct 2026

Believe Your Model: Distribution-Guided Confidence Calibration

arXiv 2603.03872 ↗First posted 4 Mar 2026cs.LGcited by 2

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

Abstract (arXiv)

Large Reasoning Models have demonstrated remarkable performance with the advancement of test-time scaling techniques, which enhances prediction accuracy by generating multiple candidate responses and selecting the most reliable answer. While prior work has analyzed that internal model signals like confidence scores can partly indicate response correctness and exhibit a distributional correlation with accuracy, such distributional information has not been fully utilized to guide answer selection. Motivated by this, we propose DistriVoting, which incorporates distributional priors as another signal alongside confidence during voting. Specifically, our method (1) first decomposes the mixed confidence distribution into positive and negative components using Gaussian Mixture Models, (2) then applies a reject filter based on positive/negative samples from them to mitigate overlap between the two distributions. Besides, to further alleviate the overlap from the perspective of distribution itself, we propose SelfStepConf, which uses step-level confidence to dynamically adjust inference process, increasing the separation between the two distributions to improve the reliability of confidences in voting. Experiments across 16 models and 5 benchmarks demonstrate that our method significantly outperforms state-of-the-art approaches.

Key facts. Cited by 2 papers in the map (data through 13 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 33 of its 36 references that are in the map. First posted on arXiv 4 Mar 2026.

Builds on

33 of 36 references.

Closest ancestors · 11

Same direction, newest first

  1. 2025The Art of Scaling Test-Time Compute for Large Language Models1 Dec
  2. 2025Majority of the Bests: Improving Best-of-N via Bootstrapping23 Nov
  3. 2025Beyond Token Length: Step Pruner for Efficient and Accurate Reasoning in Large Language Models4 Oct
  4. 2025GPO: Learning from Critical Steps to Improve LLM Reasoning19 Sep
  5. 2025Deep Think with Confidence21 Aug
  6. 2025Every Rollout Counts: Optimal Resource Allocation for Efficient Test-Time Scaling30 May
  7. 2025Value-Guided Search for Efficient Chain-of-Thought Reasoning23 May
  8. 2025A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?31 Mar
  9. 2025Scalable Best-of-N Selection for Large Language Models via Self-Certainty25 Feb
  10. 2025s1: Simple test-time scaling31 Jan
  11. 2024Learning From Correctness Without Prompting Makes LLM Efficient Reasoner28 Mar

Other references · 22

  1. 2024The Llama 3 Herd of Modelscited by 14,486
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,376
  3. 2025Qwen3 Technical Reportcited by 8,510
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,502
  5. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,657
  6. 2023Let's Verify Step by Stepcited by 3,672
  7. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,804
  8. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,077
  9. 2025Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?cited by 964
  10. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 960
  11. 2025Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoningcited by 548
  12. 2025MathArena: Evaluating LLMs on Uncontaminated Math Competitionscited by 281
10 more
  1. 2025TTRL: Test-Time Reinforcement Learningcited by 262
  2. 2023A Survey of Confidence Estimation and Calibration in Large Language Modelscited by 255
  3. 2023Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of Free-Form Large Language Modelscited by 149
  4. 2025Reasoning with Sampling: Your Base Model is Smarter Than You Thinkcited by 86
  5. 2025Tree Search for LLM Agent Reinforcement Learningcited by 75
  6. 2024Evaluating the Evaluator: Measuring LLMs' Adherence to Task Evaluation Instructionscited by 34
  7. 2025Generalist Reward Models: Found Inside Large Language Modelscited by 34
  8. 2024Confidence Calibration and Rationalization for LLMs via Multi-Agent Deliberationcited by 26
  9. 2025Revisiting Uncertainty Estimation and Calibration of Large Language Modelscited by 23
  10. 2025Scheduling Your LLM Reinforcement Learning with Reasoning Treescited by 11

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026CQC-RAG: Robust Retrieval-Augmented Generation via Cross-Query Consistencycited by 1
  2. 2026From the Inside Out: Progressive Distribution Refinement for Confidence Calibrationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Inference Time Optimization with Confidence Dynamics
  2. 2026VecCISC: Improving Confidence-Informed Self-Consistency with Reasoning Trace Clustering and Candidate Answer Selection
  3. 2026Calibration, Not Answer Selection: Distilling Internal Confidence in Reasoning Models
  4. 2026Claim-Level Confidence Calibration for Reliable Decision Making with Large Language Models
  5. 2025Confidence Improves Self-Consistency in LLMs
  6. 2025Let the Model Distribute Its Doubt: Confidence Estimation through Verbalized Probability Distribution
  7. 2026Latent space bias directions in LLMs capture confidence, not fairness
  8. 2025Distribution-Calibrated Inference Time Compute for Thinking LLM-as-a-Judge