Metraipolis › archive › dc_67307edf01de
Q-CueGraph: Query-Conditioned Visual Evidence Graphs for Multimodal Reasoning — arXiv:2608.04452
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.04452
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T11:14:20.809Z
- SHA-256 of the archived text
ae42e59c30cc155e4f337fc42402ca2eb530dbdca6b94aa72c97a979b2f516af- Stable URL of this page
https://metraipolis.com/documento/dc_67307edf01de
How to cite
Q-CueGraph: Query-Conditioned Visual Evidence Graphs for Multimodal Reasoning — arXiv:2608.04452 — arXiv (en), https://arxiv.org/abs/2608.04452 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_67307edf01de (Metraipolis archive dc_67307edf01de, sha256 ae42e59c30cc155e4f337fc42402ca2eb530dbdca6b94aa72c97a979b2f516af)
BibTeX
@misc{dc67307edf01de,
title = {Q-CueGraph: Query-Conditioned Visual Evidence Graphs for Multimodal Reasoning — arXiv:2608.04452},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.04452},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_67307edf01de; sha256 ae42e59c30cc155e4f337fc42402ca2eb530dbdca6b94aa72c97a979b2f516af},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.04452. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multimodal large language models (MLLMs) can miss fine details in a full image that they recognize in a closer view. Recovering this evidence requires deciding where to look and how much surrounding context to retain. We present Q-CueGraph, a query-conditioned evidence acquisition method for frozen MLLMs. For text-rich images, it builds a reusable graph of OCR lines and layout relations. Each question activates anchors, expands them into contextual regions, and selects candidates for a single observation window. Query-conditioned object detections support natural-image search through the same region-selection and composition interface. A lightweight candidate scorer further learns which observations support correct answers from frozen-reader feedback and training answers, without evidence-box supervision. Across six benchmarks, we examine the roles of query conditioning, evidence composition, and learned answerability. With Qwen2.5-VL-7B, Q-CueGraph raises V*Bench accuracy from 0.696 to 0.832 using 19.1% of source-image area, and retains 92% of full-image ANLS on InfographicVQA using about half the image area. The analyses show that useful evidence depends on both its relevance to the question and the context available to the reader. Q-CueGraph makes these choices explicit before answer generation. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-06 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2608.04452 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-31T11:02:13.514Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).