mlsanityData to 7 Oct 2026

CGES: Confidence-Guided Early Stopping for Efficient and Accurate Self-Consistency

arXiv 2511.02603 ↗First posted 4 Nov 2025cs.CLcited by 6

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

Abstract (arXiv)

Large language models (LLMs) are often queried multiple times at test time, with predictions aggregated by majority vote. While effective, this self-consistency (Wang et al., 2023) strategy requires a fixed number of calls and fails when the correct answer is infrequent. We introduce Confidence-Guided Early Stopping (CGES), a Bayesian framework that forms posteriors over candidate answers and adaptively halts sampling once one answer accumulates enough posterior mass. We prove guarantees in both an ideal calibrated regime and a realistic noisy-confidence regime under a directional drift condition. Averaged over five reasoning benchmarks, CGES reduces the average number of calls by 58% on average (from 16.0 to 6.7) while matching its accuracy within 0.4 percentage points of self-consistency.

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 12 of its 39 references that are in the map. First posted on arXiv 4 Nov 2025.

Builds on

12 of 39 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Efficient Test-Time Scaling via Self-Calibration25 Feb
  2. 2025s1: Simple test-time scaling31 Jan
  3. 2022Language Models (Mostly) Know What They Know11 Jul

Other references · 9

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  3. 2024Qwen2.5 Technical Reportcited by 4,713
  4. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  5. 2025The Lessons of Developing Process Reward Models in Mathematical Reasoningcited by 402
  6. 2022HyperTree Proof Search for Neural Theorem Provingcited by 192
  7. 2024Tree Search for Language Model Agentscited by 148
  8. 2024Divide, Reweight, and Conquer: A Logit Arithmetic Approach for In-Context Learningcited by 9
  9. 2025Dist2ill: Distributional Distillation for One-Pass Uncertainty Estimation in Large Language Modelscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 6.

6 papers.

  1. 2026Prune as You Generate: Online Rollout Pruning for Faster and Better RLVRcited by 8
  2. 2026Optimal Bayesian Stopping for Efficient Inference of Consistent LLM Answerscited by 7
  3. 2026CITE: Anytime-Valid Statistical Inference in LLM Self-Consistencycited by 2
  4. 2026MARS: Margin-Adversarial Risk-controlled Stopping for Parallel LLM Test-time Scalingcited by 2
  5. 2026A Single Revision Step Improves Token-Efficient LLM Reasoningcited by 0
  6. 2026Measuring Collapse and Correction in Homogeneous-Panel LLM Debatecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Sample Smart, Not Hard: Correctness-First Decoding for Better Reasoning in LLMs
  2. 2025Certified Self-Consistency: Statistical Guarantees and Test-Time Training for Reliable Reasoning in LLMs
  3. 2026ORCE: Order-Aware Alignment of Verbalized Confidence in Large Language Models
  4. 2026Early Stopping for Large Reasoning Models via Confidence Dynamics
  5. 2026CALIBER: Calibrating Confidence Before and After Reasoning in Language Models
  6. 2026Inference Time Optimization with Confidence Dynamics
  7. 2026From Drift to Coherence: Stabilizing Beliefs in LLMs
  8. 2025Robust Uncertainty Quantification for Self-Evolving Large Language Models via Continual Domain Pretraining