{"ID":2857515,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.09266","arxiv_id":"2510.09266","title":"CFVBench: A Comprehensive Video Benchmark for Fine-grained Multimodal Retrieval-Augmented Generation","abstract":"Multimodal Retrieval-Augmented Generation (MRAG) enables Multimodal Large Language Models (MLLMs) to generate responses with external multimodal evidence, and numerous video-based MRAG benchmarks have been proposed to evaluate model capabilities across retrieval and generation stages. However, existing benchmarks remain limited in modality coverage and format diversity, often focusing on single- or limited-modality tasks, or coarse-grained scene understanding. To address these gaps, we introduce CFVBench, a large-scale, manually verified benchmark constructed from 599 publicly available videos, yielding 5,360 open-ended QA pairs. CFVBench spans high-density formats and domains such as chart-heavy reports, news broadcasts, and software tutorials, requiring models to retrieve and reason over long temporal video spans while maintaining fine-grained multimodal information. Using CFVBench, we systematically evaluate 7 retrieval methods and 14 widely-used MLLMs, revealing a critical bottleneck: current models (even GPT5 or Gemini) struggle to capture transient yet essential fine-grained multimodal details. To mitigate this, we propose Adaptive Visual Refinement (AVR), a simple yet effective framework that adaptively increases frame sampling density and selectively invokes external tools when necessary. Experiments show that AVR consistently enhances fine-grained multimodal comprehension and improves performance across all evaluated MLLMs","short_abstract":"Multimodal Retrieval-Augmented Generation (MRAG) enables Multimodal Large Language Models (MLLMs) to generate responses with external multimodal evidence, and numerous video-based MRAG benchmarks have been proposed to evaluate model capabilities across retrieval and generation stages. However, existing benchmarks remai...","url_abs":"https://arxiv.org/abs/2510.09266","url_pdf":"https://arxiv.org/pdf/2510.09266v1","authors":"[\"Kaiwen Wei\",\"Xiao Liu\",\"Jie Zhang\",\"Zijian Wang\",\"Ruida Liu\",\"Yuming Yang\",\"Xin Xiao\",\"Xiao Sun\",\"Haoyang Zeng\",\"Changzai Pan\",\"Yidan Zhang\",\"Jiang Zhong\",\"Peijin Wang\",\"Yingchao Feng\"]","published":"2025-10-10T11:05:37Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[\"RAG\",\"Large Language Model\",\"Language Model\"]","has_code":false}