# The Confidence Paradox: Can LLM Know When It's Wrong

Source: [The Confidence Paradox: Can LLM Know When It's Wrong](<https://www.sota2.com/research/paper/68fa79b11e209131cea7eee3>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-30T05:01:12.438Z

Authors: Sahil Tripathi, Md Tabrez Nafis, Imran Hussain, Jiechao Gao

Published: 2025-06-30T02:06:54.000Z

## Abstract

Document Visual Question Answering (DocVQA) models often produce overconfident or ethically misaligned responses, especially under uncertainty. Existing models like LayoutLMv3, UDOP, and DONUT focus on accuracy but lack ethical calibration. We propose HonestVQA, a model-agnostic, self-supervised framework that aligns model confidence with correctness using weighted loss and contrastive learning. We introduce two new metrics Honesty Score (H-Score) and Ethical Confidence Index (ECI)-to evaluate ethical alignment. HonestVQA improves accuracy and F1 by up to 4.3% across SpDocVQA, InfographicsVQA, and SROIE datasets, while reducing overconfidence. It also generalizes well across domains, achieving 78.9% accuracy and 76.1% F1-score.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2506.23464>)

[PDF](<https://arxiv.org/pdf/2506.23464.pdf>)
