analiza aiqbench
This paper introduces CrossModalQA, a new benchmark designed to evaluate multimodal retrieval-augmented generation (RAG) systems on complex, open-domain tasks involving both text and images. By focusing on multi-hop reasoning across heterogeneous corpora, it addresses critical gaps in current evaluation methods where models often fail to construct complete evidence chains. The findings highlight that effective cross-modal retrieval is more vital for accuracy than simply scaling generator models, identifying multi-image reasoning as a key bottleneck for future system development.
CrossModalQA: A Cross-modal and Multi-hop Benchmark for Multimodal Retrieval-augmented Generation
arXiv:2609.05518v1 Announce Type: cross Abstract: Despite the strong capabilities of multimodal large language models (MLLMs), their parametric knowledge remains incomplete and difficult to update, motivating multimodal retrieval-augmented generation (RAG) to ground responses in external text and images. However, existing benchmarks face two major limitations: (i) they typically emphasize single-hop retrieval or reasoning over a small set of provided contexts rather than open-domain evidence discovery; and (ii) they provide fragmented coverage of cross-modal reasoning paths, leaving complex multi-hop and multi-image reasoning underexplored. In this paper, we introduce CrossModalQA, an open-domain benchmark for evaluating multimodal retrieval and reasoning over heterogeneous corpora. CrossModalQA contains 1,863 question-answer pairs constructed from 4,987 Wikipedia articles and 4,431 Wikimedia Commons images. It covers five complementary reasoning paths: vision-to-text, text-to-vision, vision-to-text-to-vision, multi-image intersection, and image-set reasoning. Every question requires retrieving and composing distributed textual and visual evidence, with an average reasoning depth of 3.50 hops. We construct the benchmark through multimodal knowledge graph-guided subgraph sampling and apply rule-based consistency checking and LLM verification to ensure multimodal dependence and traceable evidence. Extensive experiments demonstrate that existing multimodal RAG systems struggle to recover complete evidence chains and can underperform closed-book models when incomplete retrieval introduces distracting context. Further analysis reveals that complete cross-modal retrieval contributes more to answer accuracy than generator scaling, while multi-image retrieval and reasoning remain the primary bottlenecks limiting end-to-end performance.
Abstract:Despite the strong capabilities of multimodal large language models (MLLMs), their parametric knowledge remains incomplete and difficult to update, motivating multimodal retrieval-augmented generation (RAG) to ground responses in external text and images. However, existing benchmarks face two major limitations: (i) they typically emphasize single-hop retrieval or reasoning over a small set of provided contexts rather than open-domain evidence discovery; and (ii) they provide fragmented coverage of cross-modal reasoning paths, leaving complex multi-hop and multi-image reasoning underexplored. In this paper, we introduce CrossModalQA, an open-domain benchmark for evaluating multimodal retrieval and reasoning over heterogeneous corpora. CrossModalQA contains 1,863 question-answer pairs constructed from 4,987 Wikipedia articles and 4,431 Wikimedia Commons images. It covers five complementary reasoning paths: vision-to-text, text-to-vision, vision-to-text-to-vision, multi-image intersection, and image-set reasoning. Every question requires retrieving and composing distributed textual and visual evidence, with an average reasoning depth of 3.50 hops. We construct the benchmark through multimodal knowledge graph-guided subgraph sampling and apply rule-based consistency checking and LLM verification to ensure multimodal dependence and traceable evidence. Extensive experiments demonstrate that existing multimodal RAG systems struggle to recover complete evidence chains and can underperform closed-book models when incomplete retrieval introduces distracting context. Further analysis reveals that complete cross-modal retrieval contributes more to answer accuracy than generator scaling, while multi-image retrieval and reasoning remain the primary bottlenecks limiting end-to-end performance.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2609.05518 [cs.CV]
(or arXiv:2609.05518v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2609.05518 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 319 cuvinte
Subiecte conexe
Multi-Vector (Late Interaction) Embedding Models with Sentence Transformers
Hugging Face · 18 aug. 2026
RADD: Retrieval-Augmented Discrete Diffusion for Multi-Modal Knowledge Graph Completion
arXiv · 21 iul. 2026
GraMRAG: Orchestrating Multi-Agent Multi-Step Reasoning via Graph Memory with Reinforcement Learning
arXiv · 16 sept. 2026
Hypergraph-based Multimodal Retrieval-Augmented Generation with Incremental Refinement
arXiv · 18 aug. 2026