Losing Visual Needles in Image Haystacks: Vision Language Models are Easily Distracted in Short and Long Contexts

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Sharma, Aditya, Saxon, Michael, Wang, William Yang
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916422644400128
author Sharma, Aditya
Saxon, Michael
Wang, William Yang
author_facet Sharma, Aditya
Saxon, Michael
Wang, William Yang
contents We present LoCoVQA, a dynamic benchmark generator for evaluating long-context extractive reasoning in vision language models (VLMs). LoCoVQA augments test examples for mathematical reasoning, VQA, and character recognition tasks with increasingly long visual contexts composed of both in-distribution and out-of-distribution distractor images. Across these tasks, a diverse set of VLMs rapidly lose performance as the visual context length grows, often exhibiting a striking logarithmic decay trend. This test assesses how well VLMs can ignore irrelevant information when answering queries -- a task that is quite easy for language models (LMs) in the text domain -- demonstrating that current state-of-the-art VLMs lack this essential capability for many long-context applications.
format Preprint
id arxiv_https___arxiv_org_abs_2406_16851
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Losing Visual Needles in Image Haystacks: Vision Language Models are Easily Distracted in Short and Long Contexts
Sharma, Aditya
Saxon, Michael
Wang, William Yang
Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
We present LoCoVQA, a dynamic benchmark generator for evaluating long-context extractive reasoning in vision language models (VLMs). LoCoVQA augments test examples for mathematical reasoning, VQA, and character recognition tasks with increasingly long visual contexts composed of both in-distribution and out-of-distribution distractor images. Across these tasks, a diverse set of VLMs rapidly lose performance as the visual context length grows, often exhibiting a striking logarithmic decay trend. This test assesses how well VLMs can ignore irrelevant information when answering queries -- a task that is quite easy for language models (LMs) in the text domain -- demonstrating that current state-of-the-art VLMs lack this essential capability for many long-context applications.
title Losing Visual Needles in Image Haystacks: Vision Language Models are Easily Distracted in Short and Long Contexts
topic Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.16851