LLM Uncertainty Estimation

LLM: Large Language Model

Large language model uncertainty estimation studies how to tell when an answer, reasoning step, or action may be unreliable, and whether the cause is missing knowledge, ambiguous input, or random variation. Methods include sampling and semantic agreement, token probabilities and entropy, confidence calibration, probes of hidden representations, and Bayesian or evidential models; conformal prediction adds statistical guarantees for abstention or prediction sets. These signals are used to decide when to answer, ask for clarification, retrieve evidence, spend more computation, or defer to a person, with tests spanning factual QA, reasoning, medicine, code, and multimodal tasks. A central challenge is that no signal works reliably across settings: confidence can be miscalibrated or reflect commitment rather than correctness, and agreement may preserve shared errors, so methods need task-specific validation and robust evaluation under ambiguity and distribution shift.

Latest papers

403 papers38 papers in the last four weeks, up 9% on the four weeks before. 0.2% of all new papers.

Neighborhood Elites: Quality-Diversity Search for Uncertainty Quantification in Question Answering
Oct 29, 2026Mesut Ceylan, Arda Baris Basaran, Elgar Fleisch +1Uncertainty Quantification LLM Uncertainty Estimation
ETH Zurich, Zurich, Switzerland · University of St.Gallen, St.Gallen, Switzerland
Distilling the Shape of Uncertainty: What On-Policy Distillation Learns and How to Predict It
Oct 28, 2026Shiping Yang, Chuanqi Tang, Hossein Hajimirsadeghi +3Language Model Distillation On-Policy Distillation
Simon Fraser University · Borealis AI
What Hallucinations Have We Measured? Semantic Illusion, Confabulation, and Expressed Uncertainty
Oct 28, 2026Hazel H. Kim, Jakub Binkowski, Ekaterina Fadeeva +22Hallucination in Language Models LLM Hallucination Detection
University of Oxford · Wroclaw University of Science and Technology · ETH Zurich +14
Detect Before You Disambiguate: Perturbation-Based Input Ambiguity Detection in LLMs
Oct 26, 2026Tiejin Chen, Xiaoou Liu, Michael Clifford +5LLM Uncertainty Estimation
Arizona State University · Toyota InfoTech Labs
Hearsay or Fact? How Framing and Elicitation Shape LLM Probability Updates
Oct 20, 2026An-Sheng JhangProbability Elicitation from Language Models Belief Updating in LLMs
Dongguan University of Technology
Measuring Mechanistic Uncertainty in Large Language Models via Causal Robustness
Oct 15, 2026Aneesh Komanduri, Majid Afshar, Timothy A. Miller +1LLM Uncertainty Estimation Causal Interventions in Language Models
University of Colorado Anschutz Medical Campus · University of Wisconsin-Madison · Harvard Medical School +1
LEASH: Uncertainty-Aware Weak-to-Strong Escalation and Oversight for Computer-Use Agents
Oct 15, 2026Qingni Wang, Changdae Oh, Yuzhe Yang +4Computer-Use Agents AI Agent Monitoring
Simular · University of Wisconsin–Madison
GraphACM: Graph-based Action Consistency Measurement for Uncertainty Quantification in LLM Agents
Oct 15, 2026Yaxin Ma, Wenbo Zhang, Satya Krishna Pothapragada +1Behavioral Consistency of LLM Agents LLM Agent Evaluation
University of Florida, Gainesville, FL, USA · University of California, Irvine, CA, USA
Rubric Judgments Lack Uncertainty: Conformal Weighting for LLM Evaluation
Oct 15, 2026Licheng Pan, Haocheng Yang, Haoxuan Li +7Language Model Generation Evaluation Rubric-Based Evaluation
Zhejiang University · Xiaohongshu Inc · National University of Singapore +3
Argument Graph Uncertainty: Quantifying Uncertainty from the Logical Structure of Reasoning Chains
Oct 14, 2026Simon Baur, Sumire Honda, Wojciech Samek +1LLM Uncertainty Estimation Uncertainty Calibration
Fraunhofer Heinrich-Hertz-Institut 10587 Berlin, Germany
What Is a Clarification? Estimating Model-Dependent Input Ambiguity
Oct 14, 2026Jingqin Tang, Tim Tomov, Sebastian Schmidt +2LLM Uncertainty Estimation
School of Computation, Information & Technology, Technical University of Munich · Munich Data Science Institute · Munich Center for Machine Learning
Gray-Box Hallucination Detection in LLMs with Token Identities
Oct 14, 2026Shaked Goldshtein, Yitzchak Grunbaum, Fabrizio Frasca +2LLM Hallucination Detection LLM Uncertainty Estimation
Technion
ProbabilityBench: Model Evaluation and Calibration with Exact Ground-Truth Uncertainty
Oct 12, 2026Yuyang Dai, Yuting Tang, Futoshi Futami +1LLM Uncertainty Estimation Probability Calibration
Florida State University · The University of Tokyo · RIKEN AIP +1
Autoregressive Decoding as a Martingale: Characterizing the Stochastic Dynamics of LLM Generation
Oct 12, 2026Xiahui Zhu, Zhenyi Qi, Wanwei Liu +4Autoregressive Decoding Autoregressive Language Modeling
College of Computer, National University of Defense Technology, Changsha, Hunan, China
RL-ARC: Calibrating Large Reasoning Models via Reasoning-guided Uncertainty
Oct 8, 2026Gukhyeon Lee, SangKeun LeeLanguage Model Calibration LLM Uncertainty Estimation
Department of Artificial Intelligence, Korea University, Seoul, Republic of Korea · Department of Computer Science and Engineering, Korea University, Seoul, Republic of Korea
Characterizing Overconfident Failure in LLM-Based Code Generation
Oct 8, 2026Ravishka Rathnasuriya, Wei YangLLM Code Generation Confidence Estimation in Language Models
Institute of Systems for Advanced Computing, Fudan University, Shanghai, China
U-Space: Uncovering When and Why Uncertainty Arises in Language Models
Oct 6, 2026Tobias Braun, Nils Loose, Alexander Herzog +4LLM Uncertainty Estimation Confidence Estimation in Language Models
Technische Universität Darmstadt · University College London · Universität zu Lübeck +2
Sequential Probabilistic Uncertainty Estimation for Parallel Multi-Agent Reasoning Systems
Oct 6, 2026Tunyu Zhang, Zihao Zhao, Yusong Zhao +5LLM Uncertainty Estimation Multi-Agent LLM Systems
Rutgers University · University of Illinois Urbana-Champaign · Vanderbilt University +1
Evaluating Escalation Signals for LLM Routing: Targets, Controls, and Five Ways to Fool Yourself
Oct 5, 2026Ramin Pishehvar, Andrea Morandi, Mahesh ViswanathanLLM Routing Language Model Cascading
Cisco Systems, Inc., San Jose, CA, USA
Conformal Prediction with Paraphrase-Aware Scoring for LLM Uncertainty Quantification
Oct 3, 2026Jiayi Xin, Evan Qiang, Zihan Zhu +3Confidence Estimation in Language Models LLM Uncertainty Estimation
University of Pennsylvania
Trajectory-Derived Confidence for Reliable, Resource-Aware Clinical Text-to-SQL Agents
Oct 2, 2026Mincheol Daniel Song, Joshua Ward, Jake Jung +1LLM Agent Reliability Selective Prediction
University of California, Los Angeles
Single-Pass Uncertainty Heads for Claim-Level Hallucination Detection in Persian Medical Language Models
Oct 2, 2026Mehrdad Ghassabi, Pedram Rostami, Hamidreza Baradaran Kashani +2Hallucination Detection LLM Hallucination Detection
Faculty of Computer Engineering University of Isfahan Isfahan, Iran · School of Electrical and Computer Engineering University of Tehran Tehran, Iran · School of Computer Science University of Windsor Windsor, Canada +1
Revealing Epistemic Uncertainty in MLLMs via Causal-Invariant Masking
Oct 2, 2026Haoyang Luo, Linwei Tao, Jie Gui +4LLM Uncertainty Estimation Epistemic Uncertainty
Computer Science City University of Hong Kong · Machine Learning Research Apple · Cyber Science and Engineering Southeast University +2
Verbalized and Internal Probabilities Are Coupled in Large Language Models
Sep 30, 2026Sinead Williamson, Jiaxuan Li, Nick Foti +2Confidence Estimation in Language Models LLM Uncertainty Estimation
Apple
Bayesian Fine-tuning Yields Language Models that are as Bayesian as their Beliefs Allow
Sep 30, 2026Polina Tsvilodub, Andreas Waldis, Linlu Qiu +2Fine-Tuning LLM Fine-Tuning
University of Tübingen · Massachusetts Institute of Technology · New York University
Lingtai: What Concept Geometry Reveals--and Does Not Reveal--About LLM Inference
Sep 30, 2026Jiangang ChenLLM Inference LLM Interpretability
Chengdu Beiluoshimen Technology Co., Ltd.
Referential Uncertainty in Human--AI Collaboration
Sep 30, 2026Christian Poelitz, Finale Doshi-Velez, Siân LindleyLLM Uncertainty Estimation Human-AI Decision Making
Microsoft Research, Cambridge, UK · Harvard University, Cambridge, MA, USA
Also Small Models Can Reasonably Self-Evaluate Their Confidence
Sep 30, 2026Idil Kapikiran, Thomas Decker, Thomas RunklerConfidence Estimation in Language Models Small Language Models
Technical University of Munich · Siemens AG · LMU Munich +1
Probability is Not Enough: Exploring and Counting Divergent Tokens for Reasoning Uncertainty Quantification in LLMs
Sep 29, 2026Feiyang Li, Shengjing Liu, Qi Zhan +7Uncertainty Quantification Confidence Estimation in Language Models
College of Computer Science and Software Engineering, Shenzhen University · Tsinghua University · Behavioral and Spatial AI Lab, Peking University & Tongji University
HARISSA: Inference-Time Self-Checks for Efficient and Safe Local Language Model Deployment
Sep 29, 2026Kenan Alkiek, Moontae Lee, David Jurgens +1On-Device Language Model Inference Selective Prediction
School of Information University of Michigan Ann Arbor, MI 48109, USA · SNU-LG AI Research Center Seoul, South Korea

Oct 29, 2026J/K move · Enter open · S save

Neighborhood Elites: Quality-Diversity Search for Uncertainty Quantification in Question Answering

Mesut Ceylan, Arda Baris Basaran, Elgar Fleisch, Filipe Barata

ETH Zurich, Zurich, Switzerland · University of St.Gallen, St.Gallen, Switzerland

Uncertainty Quantification LLM Uncertainty Estimation Language Model Error Detection Question Answering Quality-Diversity Optimization

Open paper AskPDF arXiv

Large language models (LLMs) are increasingly capable, yet they produce wrong answers indistinguishable from correct ones. In question answering (QA), such an error propagates unnoticed into downstream tasks. Perturbation-based uncertainty quantification (UQ) addresses this by probing a model over a set of input rewrites and reading uncertainty from how much its predictions vary. Yet this set is usually built generically, by free paraphrasing, token-level perturbation, or adversarial search. Such rewrites leave two things uncontrolled that matter in QA: what kind of change is made to the context, and how explicitly the context supports the answer. When neither is controlled, the composition of the set is determined by the generator rather than by design, and the estimate inherits whatever that generator happens to produce. Quality-diversity (QD) search is built to cover a predefined exploration space, so we recast input perturbation as a QD problem and propose neighborhood elites (NE), a neighborhood constructor that spreads answer-preserving rewrites across two axes, the kind of edit and the explicitness of support. We compare NE against three other neighborhood constructors, holding the editor, QA model, uncertainty measure and search objective fixed, and against three scores read from the original context. Across five extractive-QA datasets and five instruction-tuned LLMs, NE improves failure detection against every other constructor in at least 24 of the 25 dataset-model pairs, reaching 0.726 AUROC and 0.211 AUPRC lift over the base rate. Additionally, NE detects failure better than semantic entropy computed over decoder samples, by 0.074 AUROC and 0.062 AUPRC lift, and this advantage widens as the neighborhood grows.