mlsanityData to 7 Oct 2026

Self-Evaluating LLMs for Multi-Step Tasks: Stepwise Confidence Estimation for Failure Detection

arXiv 2511.07364 ↗First posted 10 Nov 2025cs.LG cs.AI cs.CLcited by 6

Not on the map. Why?

Abstract (arXiv)

Reliability and failure detection of large language models (LLMs) is critical for their deployment in high-stakes, multi-step reasoning tasks. Prior work explores confidence estimation for self-evaluating LLM-scorer systems, with confidence scorers estimating the likelihood of errors in LLM responses. However, most methods focus on single-step outputs and overlook the challenges of multi-step reasoning. In this work, we extend self-evaluation techniques to multi-step tasks, testing two intuitive approaches: holistic scoring and step-by-step scoring. Using two multi-step benchmark datasets, we show that stepwise evaluation generally outperforms holistic scoring in detecting potential errors, with up to 15% relative increase in AUC-ROC. Our findings demonstrate that self-evaluating LLM systems provide meaningful confidence estimates in complex reasoning, improving their trustworthiness and providing a practical framework for failure detection.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 22 of its 26 references that are in the map. First posted on arXiv 10 Nov 2025.

Builds on

22 of 26 references.

References · 22

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  3. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  4. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 946
  5. 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 733
  6. 2018CoQA: A Conversational Question Answering Challengecited by 674
  7. 2025Multi-Agent Collaboration Mechanisms: A Survey of LLMscited by 424
  8. 2023A Survey of Confidence Estimation and Calibration in Large Language Modelscited by 253
  9. 2016Professor Forcing: A New Algorithm for Training Recurrent Networkscited by 214
  10. 2025Scalable Best-of-N Selection for Large Language Models via Self-Certaintycited by 182
  11. 2024A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systemscited by 170
  12. 2022Multi-hop Question Answeringcited by 64
10 more
  1. 2025Rewarding Doubt: A Reinforcement Learning Approach to Calibrated Confidence Expression of Large Language Modelscited by 48
  2. 2024Think Twice Before Trusting: Self-Detection for Large Language Models through Comprehensive Answer Reflectioncited by 40
  3. 2024InternalInspector I2: Robust Confidence Estimation in LLMs through Internal Statescited by 29
  4. 2024Enhancing Confidence Expression in Large Language Models Through Learning from Past Experiencecited by 21
  5. 2024Large Language Model Confidence Estimation via Black-Box Accesscited by 14
  6. 2025Language Models Prefer What They Know: Relative Confidence Estimation via Confidence Preferencescited by 12
  7. 2024Graph-based Confidence Calibration for Large Language Modelscited by 10
  8. 2025Reward Models Identify Consistency, Not Causalitycited by 6
  9. 2025Mind the Generation Process: Fine-Grained Confidence Estimation During LLM Generationcited by 5
  10. 2024Confidence Estimation for LLM-Based Dialogue State Trackingcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 6.

6 papers.

  1. 2026Critic Experience Bank: Self-Evolving Step-Level Confidence Estimation for LLM Agentscited by 2
  2. 2026iFlip: Iterative Feedback-driven Counterfactual Example Refinementcited by 1
  3. 2026The Confident Liar: Diagnosing Multi-Agent Debate with Log-Probabilities and LLM-as-Judgecited by 1
  4. 2026Early-Token Confidence Predicts Reasoning Quality in Multi-Agent LLM Debatecited by 1
  5. 2026Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Studycited by 0
  6. 2026When Do Models Admit They Are Wrong? Failure Disclosure Is Unstable Under Reinforcement Learningcited by 0