# Overconfidence is Key: Verbalized Uncertainty Evaluation in Large Language and Vision-Language Models

arXiv: https://arxiv.org/abs/2405.02917 · HTML: https://mlsanity.com/paper/2405.02917/

## Key facts

- Cited by 54 papers in the map (data through 14 Oct 2026).
- Placed in Hallucination mitigation in LVLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 7 of its 18 references that are in the map.
- First posted on arXiv 5 May 2024.

## Placed in

- Multimodal LLMs / vision-language models / [Hallucination mitigation in LVLMs](https://mlsanity.com/mllm/hallucination-mitigation-in-lvlms/)

## Builds on (most cited first, 7 of 7)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs](https://mlsanity.com/paper/2306.13063/) (2023)
- [How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering](https://mlsanity.com/paper/2012.00955/) (2020)
- [I Find Your Lack of Uncertainty in Computer Vision Disturbing](https://mlsanity.com/paper/2104.08188/) (2021)

## Built on by (most cited first, 30 of 54)

- [A Survey on Uncertainty Quantification of Large Language Models: Taxonomy, Open Research Challenges, and Future Directions](https://mlsanity.com/paper/2412.05563/) (2024)
- [From Calibration to Collaboration: LLM Uncertainty Quantification Should Be More Human-Centered](https://mlsanity.com/paper/2506.07461/) (2025)
- [Prune 'n Predict: Optimizing LLM Decision-making with Conformal Prediction](https://mlsanity.com/paper/2501.00555/) (2024)
- [Agentic Confidence Calibration](https://mlsanity.com/paper/2601.15778/) (2026)
- [Anthropomimetic Uncertainty: What Verbalized Uncertainty in Language Models is Missing](https://mlsanity.com/paper/2507.10587/) (2025)
- [Grammars of Formal Uncertainty: When to Trust LLMs in Automated Reasoning Tasks](https://mlsanity.com/paper/2505.20047/) (2025)
- [Do Large Language Models Know What They Are Capable Of?](https://mlsanity.com/paper/2512.24661/) (2025)
- [The Illusion of Certainty: Decoupling Capability and Calibration in On-Policy Distillation](https://mlsanity.com/paper/2604.16830/) (2026)
- [Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring](https://mlsanity.com/paper/2605.30834/) (2026)
- [Calibrating Uncertainty Quantification of Multi-Modal LLMs using Grounding](https://mlsanity.com/paper/2505.03788/) (2025)
- [Closing the Confidence-Faithfulness Gap in Large Language Models](https://mlsanity.com/paper/2603.25052/) (2026)
- [CaRT: Teaching LLM Agents to Know When They Know Enough](https://mlsanity.com/paper/2510.08517/) (2025)
- [Overconfidence and Calibration in Medical VQA: Empirical Findings and Hallucination-Aware Mitigation](https://mlsanity.com/paper/2604.02543/) (2026)
- [Enhancing Trust in Large Language Models via Uncertainty-Calibrated Fine-Tuning](https://mlsanity.com/paper/2412.02904/) (2024)
- [Uncertainty in Action: Confidence Elicitation in Embodied Agents](https://mlsanity.com/paper/2503.10628/) (2025)
- [Visual Instruction Bottleneck Tuning](https://mlsanity.com/paper/2505.13946/) (2025)
- [Outlier Synthesis via Hamiltonian Monte Carlo for Out-of-Distribution Detection](https://mlsanity.com/paper/2501.16718/) (2025)
- [Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQA](https://mlsanity.com/paper/2511.11169/) (2025)
- [Instinct vs. Reflection: Unifying Token and Verbalized Confidence in Multimodal Large Models](https://mlsanity.com/paper/2604.17274/) (2026)
- [Are vision language models robust to uncertain inputs?](https://mlsanity.com/paper/2505.11804/) (2025)
- [On the Robustness of Verbal Confidence of LLMs in Adversarial Attacks](https://mlsanity.com/paper/2507.06489/) (2025)
- [Do We Really Even Need Data? A Modern Look at Drawing Inference with Predicted Data](https://mlsanity.com/paper/2512.05456/) (2025)
- [The Dilemma of Uncertainty Estimation for General Purpose AI in the EU AI Act](https://mlsanity.com/paper/2408.11249/) (2024)
- [Reflexive Guidance: Improving OoDD in Vision-Language Models via Self-Guided Image-Adaptive Concept Generation](https://mlsanity.com/paper/2410.14975/) (2024)
- [Prune-Then-Plan: Step-Level Calibration for Stable Frontier Exploration in Embodied Question Answering](https://mlsanity.com/paper/2511.19768/) (2025)
- [Pretraining with random noise for uncertainty calibration](https://mlsanity.com/paper/2412.17411/) (2024)
- [Uncertainty Quantification for Motor Imagery BCI -- Machine Learning vs. Deep Learning](https://mlsanity.com/paper/2507.07511/) (2025)
- [Similarity-as-Evidence: Calibrating Overconfident VLMs for Interpretable and Label-Efficient Medical Active Learning](https://mlsanity.com/paper/2602.18867/) (2026)
- [SIEVES: Selective Prediction Generalizes through Visual Evidence Scoring](https://mlsanity.com/paper/2604.25855/) (2026)
- [Calibrating LLMs with Semantic-level Reward](https://mlsanity.com/paper/2605.15588/) (2026)

## Abstract (arXiv)

Language and Vision-Language Models (LLMs/VLMs) have revolutionized the field of AI by their ability to generate human-like text and understand images, but ensuring their reliability is crucial. This paper aims to evaluate the ability of LLMs (GPT4, GPT-3.5, LLaMA2, and PaLM 2) and VLMs (GPT4V and Gemini Pro Vision) to estimate their verbalized uncertainty via prompting. We propose the new Japanese Uncertain Scenes (JUS) dataset, aimed at testing VLM capabilities via difficult queries and object counting, and the Net Calibration Error (NCE) to measure direction of miscalibration. Results show that both LLMs and VLMs have a high calibration error and are overconfident most of the time, indicating a poor capability for uncertainty estimation. Additionally we develop prompts for regression tasks, and we show that VLMs have poor calibration when producing mean/standard deviation and 95% confidence intervals.