Logit-Level Uncertainty Quantification in Vision-Language Models for Histopathology Image Analysis

TL;DR

This paper introduces a logit-level uncertainty quantification framework for vision-language models in histopathology, highlighting differences in model reliability and stability across various VLMs and prompts.

Contribution

It presents a novel UQ framework at the logit level for histopathology VLMs, enabling better assessment of trustworthiness and uncertainty behavior in medical image analysis.

Findings

VLMs exhibit high stochastic sensitivity and abrupt uncertainty transitions for complex prompts.

The pathology-specific PRISM model maintains near-deterministic behavior with minimal temperature effects.

Logit-level UQ effectively evaluates trustworthiness in medical image analysis.

Abstract

Vision-Language Models (VLMs) with their multimodal capabilities have demonstrated remarkable success in almost all domains, including education, transportation, healthcare, energy, finance, law, and retail. Nevertheless, the utilization of VLMs in healthcare applications raises crucial concerns due to the sensitivity of large-scale medical data and the trustworthiness of these models (reliability, transparency, and security). This study proposes a logit-level uncertainty quantification (UQ) framework for histopathology image analysis using VLMs to deal with these concerns. UQ is evaluated for three VLMs using metrics derived from temperature-controlled output logits. The proposed framework demonstrates a critical separation in uncertainty behavior. While VLMs show high stochastic sensitivity (cosine similarity (CS) $<0.71$ and $<0.84$, Jensen-Shannon divergence (JS) $<0.57$ and $<0.38$, and Kullback-Leibler divergence (KL) $<0.55$ and $<0.35$, respectively for mean values of VILA-M3-8B and LLaVA-Med v1.5), near-maximal temperature impacts ($\Delta_T \approx 1.00$), and displaying abrupt uncertainty transitions, particularly for complex diagnostic prompts. In contrast, the pathology-specific PRISM model maintains near-deterministic behavior (mean CS $>0.90$, JS $<0.10$, KL $<0.09$) and significantly minimal temperature effects across all prompt complexities. These findings emphasize the importance of logit-level uncertainty quantification to evaluate trustworthiness in histopathology applications utilizing VLMs.