mlsanityData to 7 Oct 2026

Agentic Uncertainty Reveals Agentic Overconfidence

arXiv 2602.06948 ↗First posted 6 Feb 2026cs.AI cs.LGcited by 7

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

Can AI agents predict whether they will succeed at a task? We study agentic uncertainty by eliciting success probability estimates before, during, and after task execution. All results exhibit agentic overconfidence: some agents that succeed only 22% of the time predict 77% success. Counterintuitively, pre-execution assessment with strictly less information tends to yield better discrimination than standard post-execution review, though differences are not always significant. Adversarial prompting reframing assessment as bug-finding achieves the best calibration.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 23 of its 25 references that are in the map. First posted on arXiv 6 Feb 2026.

Builds on

23 of 25 references.

References · 23

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 1,908
  3. 2022Language Models (Mostly) Know What They Knowcited by 1,783
  4. 2024A Survey on LLM-as-a-Judgecited by 1,384
  5. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 965
  6. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 795
  7. 2024LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methodscited by 451
  8. 2018AI safety via debatecited by 384
  9. 2024Debating with More Persuasive LLMs Leads to More Truthful Answerscited by 255
  10. 2024When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMscited by 230
  11. 2023AI Control: Improving Safety Despite Intentional Subversioncited by 207
  12. 2025Measuring AI Ability to Complete Long Software Taskscited by 142
11 more
  1. 2024On the Self-Verification Limitations of Large Language Models on Reasoning and Planning Taskscited by 137
  2. 2025Beyond Binary Rewards: Training LMs to Reason About Their Uncertaintycited by 107
  3. 2026Emergent Introspective Awareness in Large Language Modelscited by 73
  4. 2025Ctrl-Z: Controlling AI Agents via Resamplingcited by 40
  5. 2025Play Favorites: A Statistical Method to Measure Self-Bias in LLM-as-a-Judgecited by 35
  6. 2025Large Language Models are overconfident and amplify human biascited by 30
  7. 2025Beyond the Surface: Measuring Self-Preference in LLM Judgmentscited by 30
  8. 2024Can We Trust LLMs? Mitigate Overconfidence Bias in LLMs through Knowledge Transfercited by 19
  9. 2025Overconfidence in LLM-as-a-Judge: Diagnosis and Confidence-Driven Solutioncited by 17
  10. 2025ToolRM: Outcome Reward Models for Tool-Calling Large Language Modelscited by 16
  11. 2025Do Large Language Models Know What They Are Capable Of?cited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 7.

7 papers.

  1. 2026Position: agentic AI orchestration should be Bayes-consistentcited by 3
  2. 2026Uncertainty Decomposition for Clarification Seeking in LLM Agentscited by 3
  3. 2026Synergy: A Next-Generation General-Purpose Agent for Open Agentic Webcited by 2
  4. 2026Coupling Planning with Episodic Memory in LLM Agents for Software Issue Resolutioncited by 1
  5. 2026Online Safety Monitoring for LLMscited by 0
  6. 2026Bringing analytic rigor to agentic AI for science: The Brain Researcher platform for neuroimaging data analysiscited by 0
  7. 2026RECLAIM: Can Agents Reproduce the Claims of Machine Learning Papers?cited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Do Agents Know When They Succeed? Calibrating Agent Confidence from Internal Representations
  2. 2026Predicting Task Difficulty Without Rollouts
  3. 2026Agentic Confidence Calibration
  4. 2026The Confidence Game: Strategic Miscalibration in Human-AI Delegation
  5. 2026Agentic Uncertainty Quantification
  6. 2026On Randomness in Agentic Evals
  7. 2025Establishing Best Practices for Building Rigorous Agentic Benchmarks
  8. 2025Stochasticity in Agentic Evaluations: Quantifying Inconsistency with Intraclass Correlation