What Makes "Good" Distractors for Object Hallucination Evaluation in Large Vision-Language Models?

TL;DR

This paper introduces the HOPE benchmark, a new method for evaluating object hallucination in large vision-language models by generating highly misleading distractors, revealing their vulnerabilities more effectively than previous benchmarks.

Contribution

The paper proposes the HOPE benchmark, which uses content-aware and description-based hallucination searching to better assess LVLMs' object hallucination issues.

Findings

HOPE causes a 9-23% performance drop in state-of-the-art LVLMs.

HOPE outperforms POPE in exposing hallucination vulnerabilities.

Experimental results validate HOPE's effectiveness in rigorous hallucination assessment.

Abstract

Large Vision-Language Models (LVLMs), empowered by the success of Large Language Models (LLMs), have achieved impressive performance across domains. Despite the great advances in LVLMs, they still suffer from the unavailable object hallucination issue, which tends to generate objects inconsistent with the image content. The most commonly used Polling-based Object Probing Evaluation (POPE) benchmark evaluates this issue by sampling negative categories according to category-level statistics, \textit{e.g.}, category frequencies and co-occurrence. However, with the continuous advancement of LVLMs, the POPE benchmark has shown diminishing effectiveness in assessing object hallucination, as it employs a simplistic sampling strategy that overlooks image-specific information and restricts distractors to negative object categories only. In this paper, we introduce the Hallucination searching-based Object Probing Evaluation (HOPE) benchmark, aiming to generate the most misleading distractors (\textit{i.e.}, non-existent objects or incorrect image descriptions) that can trigger hallucination in LVLMs, which serves as a means to more rigorously assess their immunity to hallucination. To explore the image-specific information, the content-aware hallucination searching leverages Contrastive Language-Image Pre-Training (CLIP) to approximate the predictive behavior of LVLMs by selecting negative objects with the highest predicted likelihood as distractors. To expand the scope of hallucination assessment, the description-based hallucination searching constructs highly misleading distractors by pairing true objects with false descriptions. Experimental results show that HOPE leads to a precision drop of at least 9\% and up to 23\% across various state-of-the-art LVLMs, significantly outperforming POPE in exposing hallucination vulnerabilities. The code is available at https://github.com/xiemk/HOPE.