Saved in:
Bibliographic Details
Main Authors: Tan, Xichen, Ye, Yunfan, Luo, Yuanjing, Wan, Qian, Liu, Fang, Cai, Zhiping
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.08576
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910870236299264
author Tan, Xichen
Ye, Yunfan
Luo, Yuanjing
Wan, Qian
Liu, Fang
Cai, Zhiping
author_facet Tan, Xichen
Ye, Yunfan
Luo, Yuanjing
Wan, Qian
Liu, Fang
Cai, Zhiping
contents Multi-modal Large Language Models (MLLMs) capable of video understanding are advancing rapidly. To effectively assess their video comprehension capabilities, long video understanding benchmarks, such as Video-MME and MLVU, are proposed. However, these benchmarks directly use uniform frame sampling for testing, which results in significant information loss and affects the accuracy of the evaluations in reflecting the true abilities of MLLMs. To address this, we propose RAG-Adapter, a plug-and-play framework that reduces information loss during testing by sampling frames most relevant to the given question. Additionally, we introduce a Grouped-supervised Contrastive Learning (GCL) method to further enhance sampling effectiveness of RAG-Adapter through fine-tuning on our constructed MMAT dataset. Finally, we test numerous baseline MLLMs on various video understanding benchmarks, finding that RAG-Adapter sampling consistently outperforms uniform sampling (e.g., Accuracy of GPT-4o increases by 9.3 percent on Video-MME), providing a more accurate testing method for long video benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2503_08576
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RAG-Adapter: A Plug-and-Play RAG-enhanced Framework for Long Video Understanding
Tan, Xichen
Ye, Yunfan
Luo, Yuanjing
Wan, Qian
Liu, Fang
Cai, Zhiping
Computer Vision and Pattern Recognition
Multi-modal Large Language Models (MLLMs) capable of video understanding are advancing rapidly. To effectively assess their video comprehension capabilities, long video understanding benchmarks, such as Video-MME and MLVU, are proposed. However, these benchmarks directly use uniform frame sampling for testing, which results in significant information loss and affects the accuracy of the evaluations in reflecting the true abilities of MLLMs. To address this, we propose RAG-Adapter, a plug-and-play framework that reduces information loss during testing by sampling frames most relevant to the given question. Additionally, we introduce a Grouped-supervised Contrastive Learning (GCL) method to further enhance sampling effectiveness of RAG-Adapter through fine-tuning on our constructed MMAT dataset. Finally, we test numerous baseline MLLMs on various video understanding benchmarks, finding that RAG-Adapter sampling consistently outperforms uniform sampling (e.g., Accuracy of GPT-4o increases by 9.3 percent on Video-MME), providing a more accurate testing method for long video benchmarks.
title RAG-Adapter: A Plug-and-Play RAG-enhanced Framework for Long Video Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.08576