Language Model Calibration

Follow this topic Ask about this topic

Language model calibration asks whether a model’s stated or computed confidence corresponds to how often its answers are correct, so systems can decide when to answer, retrieve evidence, defer, or abstain. Researchers estimate confidence from verbal reports, token probabilities, self-consistency, or hidden representations, then improve it with post-hoc methods such as temperature scaling and isotonic regression, conformal prediction, or training methods including supervised fine-tuning and reinforcement learning. Evaluations commonly use question answering and mathematical reasoning benchmarks, and increasingly test retrieval routing, model judges, medical tasks, and performance under domain or prompt shifts. A central tension is that calibration can change with the elicitation protocol and deployment context, and well-calibrated probabilities do not necessarily rank errors well or guarantee safe decisions.

Latest papers

386 papers67 papers in the last four weeks, up 40% on the four weeks before. 0.4% of all new papers.

Assessing and Mitigating Miscalibration in LLM-Based Social Science Measurement
May 12, 2026Jinyuan Wang, Ningyuan Deng, Yi YangLanguage Model Calibration LLM Reliability
The Hong Kong University of Science and Technology
Probabilistic Calibration Is a Trainable Capability in Language Models
May 12, 2026Davide Baldelli, Sruthi Kuriakose, Maryam Hashemzadeh +2LLM Fine-Tuning Controllable Text Generation
Chandar Research Lab · Mila – Quebec AI Institute · LAMA-WeST Lab +3
ROMER: Expert Replacement and Router Calibration for Robust MoE LLMs on Analog Compute-in-Memory Systems
May 12, 2026Wenyong Zhou, Yuannuo Feng, Yizhe Chen +6Compute-in-Memory Mixture-of-Experts Language Models
The Department of Electrical and Computer Engineering, The University of Hong Kong, Hong Kong · The School of Integrated Circuit Science and Engineering, Beihang University, Beijing, China
Sampling More, Getting Less: Calibration is the Diversity Bottleneck in LLMs
May 11, 2026Amin Banayeeanzade, Qingchuan Yang, Dhruv Tarsadiya +6LLM Evaluation Language Model Decoding
♠University of Southern California · ♣Capital One
Training-Free Cultural Alignment of Large Language Models via Persona Disagreement
May 11, 2026Huynh Trung Kiet, Dao Sy Duy Minh, Tuan Nguyen +5LLM Alignment Cultural Alignment
Faculty of Information and Technology, University of Science, Vietnam National University, Ho Chi Minh City, Vietnam. · Department of Computer Science, University of Warwick, Coventry, United Kingdom. · School of Computing, Engineering and Digital Technologies, Teesside University, United Kingdom.
ConQuR: Corner Aligned Activation Quantization via Optimized Rotations for LLMs
May 11, 2026Chayne Thrash, Ali Abbasi, Soheil KolouriRotation-Based Quantization LLM Inference Acceleration
Department of Computer Science Vanderbilt University
Auditing Multimodal LLM Raters: Central Tendency Bias in Clinical Ordinal Scoring
May 11, 2026Jiaqing Zhang, Sandeep Elluri, Bhanu Cherukuvada +7LLM Auditing Multimodal Large Language Models
Department of Electrical & Computer Engineering · Department of Computer and Information Science and Engineering · Department of Clinical and Health Psychology +1
Task-Aware Calibration: Provably Optimal Decoding in LLMs
May 11, 2026Tim Tomov, Dominik Fuchsgruber, Rajeev Verma +1Language Model Decoding MBR Decoding
School of Computation, Information & Technology, Technical University of Munich · Munich Data Science Institute · Munich Center for Machine Learning +1
UCCI: Calibrated Uncertainty for Cost-Optimal LLM Cascade Routing
May 11, 2026Varun KotteCost-Aware Inference Language Model Calibration
Independent Researcher.
The Metacognitive Probe: Five Behavioural Calibration Diagnostics for LLMs
May 11, 2026Rafael C. T. OliveiraLanguage Model Calibration LLM Uncertainty Estimation
Independent Researcher
The Silent Vote: Improving Zero-Shot LLM Reliability by Aggregating Semantic Neighborhoods
May 10, 2026Sanket Badhe, Priyanka Tiwari, Deep ShahConstrained Decoding Zero-Shot Text Classification
Google
Calibrate, Don't Curate: Label-Efficient Estimation from Noisy LLM Judges
May 10, 2026Yanran LiLLM Evaluation LLM-as-a-Judge
Department of Biostatistics Columbia University
NICE FACT: Diagnosing and Calibrating VLMs in Quantitative Reasoning for Kinematic Physics
May 8, 2026Jian Lan, Zhicheng Liu, Xinpeng Wang +5VLM Evaluation Physical Reasoning
University of Munich (LMU), Germany 2Munich Center of Machine Learning3 Sichuan University
A Semantic-Sampling Framework for Evaluating Calibration in Open-Ended Question Answering
May 8, 2026Zhanliang Wang, Jiancong Xiao, Ruochen Jin +3LLM Evaluation Confidence Estimation in Language Models
University of Pennsylvania, Philadelphia, PA · Dartmouth College, Hanover, NH
Beyond Confidence: Rethinking Self-Assessments for Performance Prediction in LLMs
May 8, 2026Sree Bhattacharyya, Samarth Khanna, Leona Chen +3LLM Evaluation Confidence Estimation in Language Models
Department of Informatics and Intelligent Systems, College of Information Sciences, The Pennsylvania State University. · Department of Computer Science and Engineering, College of Engineering, The Pennsylvania State University.
Bayesian Fine-tuning in Projected Subspaces
May 8, 2026Viktar Dubovik, Patryk Marszałek, Jacek Tabor +1Bayesian Neural Networks Fine-Tuning
Jagiellonian University, Kraków, Poland · Doctoral School of Exact and Natural Sciences, Jagiellonian University, Kraków, Poland
Confidence-Aware Alignment Makes Reasoning LLMs More Reliable
May 8, 2026Kejia Chen, Jiawen Zhang, Yihong Wu +5LLM Alignment Direct Preference Optimization
Zhejiang University · Université de Montréal · Sun Yat-sen University +1
Bias and Uncertainty in LLM-as-a-Judge Estimation
May 7, 2026James FiedlerLLM Evaluation LLM-as-a-Judge
Indeed Inc.
A Reproducible Optimisation Protocol for Calibrating Prompt-Based Large Language Model Workflows in Evidence Synthesis
May 7, 2026Teo SusnjakLarge Language Model-Guided Optimization Language Model Calibration
School of Mathematical and Computational Sciences Massey University, Auckland, New Zealand
Log-Likelihood, Simpson's Paradox, and the Detection of Machine-Generated Text
May 7, 2026Tom Kempton, Viktor Drobnyi, Maeve Madigan +1AI-Generated Text Detection Language Model Calibration
Department of Mathematics University of Manchester · Risk and Security AI Lab Visa Inc.
Saliency-Aware Regularized Quantization Calibration for Large Language Models
May 7, 2026Yanlong Zhao, Xiaoyuan Cheng, Huihang Liu +6LLM Quantization Language Model Calibration
University of Science and Technology of China · University College London · 3Shanghai University of Finance and Economics +5
BitCal-TTS: Bit-Calibrated Test-Time Scaling for Quantized Reasoning Models
May 7, 2026Sai Babu Patarlapalli, Surya Teja AvvaruLLM Inference 4-Bit Quantization
Distilling Bayesian Belief States into Language Models for Auditable Negotiation
May 6, 2026Zongqi Cui, Baihan LinMulti-Agent Negotiation Opponent Modeling
Emory University · Icahn School of Medicine at Mount Sinai
EQUITRIAGE: A Fairness Audit of Gender Bias in LLM-Based Emergency Department Triage
May 5, 2026Richard J. Young, Alice M. MatthewsGender Bias in Language Models Clinical Triage
University of Nevada, Las Vegas, Lee Business School, Las Vegas, NV, USA · DeepNeuro AI, Las Vegas, NV, USA · Concorde Career Colleges, Dept. of Cardiovascular and Medical Diagnostic Sonography
Reliability-Oriented Multilingual Orthopedic Diagnosis: A Domain-Adaptive Modeling and a Conceptual Validation Framework
May 4, 2026Danish Ali, Li Xiaojian, Sundas Iqbal +1Multilingual Language Model Evaluation Healthcare
School of Computer Science, Wuhan University, China · School of Software, Nanjing University of Information Science and Technology (NUIST), China · Department of Orthopedic, Bahawal Victoria Hospital, Pakistan
What Single-Prompt Accuracy Misses: A Multi-Variant Reliability Audit of Language Models
May 3, 2026Ranit Karmakar, Jayita ChatterjeeLLM Evaluation Language Model Calibration
Harvard University, Boston, MA 02115 · Boston, MA 02134
Only Say What You Know: Calibration-Aware Generation for Long-Form Factuality
May 3, 2026Wen Luo, Guangyue Peng, Liang Wang +7Long-Form Document Generation LLM Hallucination Mitigation
State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University Research Asia · MSRA · Microsoft
Geometry-Calibrated Conformal Abstention for Language Models
Apr 30, 2026Rui Xu, Yi Chen, Sihong Xie +1Uncertainty Quantification Selective Prediction
Information Hub, AI Thrust Hong Kong University of Science and Technology (Guangzhou) Guangzhou, Guangdong, China
LLMs Capture Emotion Labels, Not Emotion Uncertainty: Distributional Analysis and Calibration of Human-LLM Judgment Gaps
Apr 30, 2026Keito Inoshita, Xiaokang Zhou, Akira Kawai +1LLM Evaluation Annotator Disagreement
Faculty of Business and Commerce, Kansai University, Osaka, Japan · Faculty of Business Data Science, Kansai University, Osaka, Japan · RIKEN Center for Advanced Intelligence Project, Tokyo, Japan +2
Authority Inversion in LLM-Mediated Ubiquitous Systems: When Models Trust Users Over Sensors
Apr 28, 2026Long Zhang, Zi-bo Qin, Wei-neng ChenKnowledge Conflicts in Language Models LLM Auditing
School of Computer Science and Engineering, South China University of Technology, Guangzhou, China

May 12, 2026J/K move · Enter open · S save

Assessing and Mitigating Miscalibration in LLM-Based Social Science Measurement

Jinyuan Wang, Ningyuan Deng, Yi Yang

The Hong Kong University of Science and Technology

Language Model Calibration LLM Reliability LLM Uncertainty Estimation

Open paper AskPDF arXiv

Large language models (LLMs) are increasingly used in social science as scalable measurement tools for converting unstructured text into variables that can enter standard empirical designs. Measurement validity demands more than high average accuracy, which requires well calibrated confidence that faithfully reflects the empirical probability of each measurement being correct. This paper studies the model miscalibration in LLM-based social science measurement. We begin with a case study on FOMC and show that confidence based filtering can change downstream regression estimates when LLM confidence is miscalibrated. We then audit calibration across 14 social science constructs covering both proprietary models, including GPT-5-mini, DeepSeek-V3.2, and open source models. Across tasks and model families, reported confidence is poorly aligned with tolerance-based correctness. As a simple mitigation, we propose a soft label distillation pipeline for calibrating Bert with LLM. The method converts an LLM score and its verbalized confidence into a soft target distribution, then trains a smaller discriminative classifier on encoder models for these targets. Averaged across datasets, this approach reduces ECE by 43.2% and Brier by 34.0%. These results suggest that LLM-based social science pipelines should treat calibration as part of measurement validity, rather than as an optional post-processing concern.