Evaluating and Analyzing Relationship Hallucinations in Large Vision-Language Models
- URL: http://arxiv.org/abs/2406.16449v4
- Date: Thu, 18 Jul 2024 04:39:29 GMT
- Title: Evaluating and Analyzing Relationship Hallucinations in Large Vision-Language Models
- Authors: Mingrui Wu, Jiayi Ji, Oucheng Huang, Jiale Li, Yuhang Wu, Xiaoshuai Sun, Rongrong Ji,
- Abstract summary: We introduce R-Bench, a novel benchmark for evaluating Vision Relationship Hallucination.
R-Bench features image-level questions that focus on the existence of relationships and instance-level questions that assess local visual comprehension.
We identify three types of relationship co-occurrences that lead to hallucinations: relationship-relationship, subject-relationship, and relationship-object.
- Score: 69.79709804046325
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: The issue of hallucinations is a prevalent concern in existing Large Vision-Language Models (LVLMs). Previous efforts have primarily focused on investigating object hallucinations, which can be easily alleviated by introducing object detectors. However, these efforts neglect hallucinations in inter-object relationships, which is essential for visual comprehension. In this work, we introduce R-Bench, a novel benchmark for evaluating Vision Relationship Hallucination. R-Bench features image-level questions that focus on the existence of relationships and instance-level questions that assess local visual comprehension. We identify three types of relationship co-occurrences that lead to hallucinations: relationship-relationship, subject-relationship, and relationship-object. The visual instruction tuning dataset's long-tail distribution significantly impacts LVLMs' understanding of visual relationships. Furthermore, our analysis reveals that current LVLMs tend to disregard visual content and overly rely on the common sense knowledge of Large Language Models. They also struggle with reasoning about spatial relationships based on contextual information.
Related papers
- Devils in Middle Layers of Large Vision-Language Models: Interpreting, Detecting and Mitigating Object Hallucinations via Attention Lens [7.806633929976787]
Hallucinations in Large Vision-Language Models (LVLMs) significantly undermine their reliability.
This paper addresses how LVLMs process visual information and whether this process causes hallucination.
We propose a simple inference-time method that adjusts visual attention by integrating information across various heads.
arXiv Detail & Related papers (2024-11-23T03:40:05Z) - Unified Triplet-Level Hallucination Evaluation for Large Vision-Language Models [22.996176483599868]
We design a unified framework to measure object and relation hallucination in Large Vision-Language Models (LVLMs) simultaneously.
Based on our framework, we introduce Tri-HE, a novel Triplet-level Hallucination Evaluation benchmark.
arXiv Detail & Related papers (2024-10-30T15:25:06Z) - Reefknot: A Comprehensive Benchmark for Relation Hallucination Evaluation, Analysis and Mitigation in Multimodal Large Language Models [13.48296910438554]
Hallucination issues persistently plagued current multimodal large language models (MLLMs)
We introduce Reefknot, a benchmark specifically targeting relation hallucinations, consisting of over 20,000 samples derived from real-world scenarios.
Our comparative evaluation across three distinct tasks revealed a substantial shortcoming in the capabilities of current MLLMs to mitigate relation hallucinations.
arXiv Detail & Related papers (2024-08-18T10:07:02Z) - VALOR-EVAL: Holistic Coverage and Faithfulness Evaluation of Large Vision-Language Models [57.43276586087863]
Large Vision-Language Models (LVLMs) suffer from hallucination issues, wherein the models generate plausible-sounding but factually incorrect outputs.
Existing benchmarks are often limited in scope, focusing mainly on object hallucinations.
We introduce a multi-dimensional benchmark covering objects, attributes, and relations, with challenging images selected based on associative biases.
arXiv Detail & Related papers (2024-04-22T04:49:22Z) - RelationVLM: Making Large Vision-Language Models Understand Visual Relations [66.70252936043688]
We present RelationVLM, a large vision-language model capable of comprehending various levels and types of relations whether across multiple images or within a video.
Specifically, we devise a multi-stage relation-aware training scheme and a series of corresponding data configuration strategies to bestow RelationVLM with the capabilities of understanding semantic relations.
arXiv Detail & Related papers (2024-03-19T15:01:19Z) - Mitigating Hallucination in Visual Language Models with Visual
Supervision [33.05550629039951]
Large vision-language models (LVLMs) suffer from hallucination a lot.
Key problem lies in its weak ability to comprehend detailed content in a multi-modal context.
In this paper, we bring more detailed vision annotations and more discriminative vision models to facilitate the training of LVLMs.
arXiv Detail & Related papers (2023-11-27T09:30:02Z) - Towards Mitigating Hallucination in Large Language Models via
Self-Reflection [63.2543947174318]
Large language models (LLMs) have shown promise for generative and knowledge-intensive tasks including question-answering (QA) tasks.
This paper analyses the phenomenon of hallucination in medical generative QA systems using widely adopted LLMs and datasets.
arXiv Detail & Related papers (2023-10-10T03:05:44Z) - Constellation: Learning relational abstractions over objects for
compositional imagination [64.99658940906917]
We introduce Constellation, a network that learns relational abstractions of static visual scenes.
This work is a first step in the explicit representation of visual relationships and using them for complex cognitive procedures.
arXiv Detail & Related papers (2021-07-23T11:59:40Z) - Visual Relationship Detection with Visual-Linguistic Knowledge from
Multimodal Representations [103.00383924074585]
Visual relationship detection aims to reason over relationships among salient objects in images.
We propose a novel approach named Visual-Linguistic Representations from Transformers (RVL-BERT)
RVL-BERT performs spatial reasoning with both visual and language commonsense knowledge learned via self-supervised pre-training.
arXiv Detail & Related papers (2020-09-10T16:15:09Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.