Metraipolis › archive › dc_ad98e677c4cc

Navigating Sparse Evidence: Agentic Visual RAG via Explicit Context Selection and Consolidation — arXiv:2609.15800

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.15800
License (as archived)
CC0-1.0
Archived
2026-10-15T22:48:06.620Z
SHA-256 of the archived text
9cd9e026747a31dfc954c039c71b3a95b7e5b7562838cdc5796b76ea25d02d45
Stable URL of this page
https://metraipolis.com/documento/dc_ad98e677c4cc

How to cite

Navigating Sparse Evidence: Agentic Visual RAG via Explicit Context Selection and Consolidation — arXiv:2609.15800 — arXiv (en), https://arxiv.org/abs/2609.15800 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ad98e677c4cc (Metraipolis archive dc_ad98e677c4cc, sha256 9cd9e026747a31dfc954c039c71b3a95b7e5b7562838cdc5796b76ea25d02d45)
BibTeX
@misc{dcad98e677c4cc,
  title = {Navigating Sparse Evidence: Agentic Visual RAG via Explicit Context Selection and Consolidation — arXiv:2609.15800},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.15800},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ad98e677c4cc; sha256 9cd9e026747a31dfc954c039c71b3a95b7e5b7562838cdc5796b76ea25d02d45},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15800. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Visual Retrieval-Augmented Generation (VRAG) empowers models to navigate and answer queries about visually rich documents by retrieving relevant page images as visual evidence and reasoning over their content. However, effectively utilizing this visual evidence is usually impeded by two main challenges. First, answer-relevant evidence is sparse and may be concentrated in a small region of one page or dispersed across multiple pages. Second, existing agentic methods often generate answers based on raw exploration trajectories or compressed textual memories rather than an explicitly organized set of supporting images, making answers susceptible to exploration noise and obscuring the evidence-backed reasoning trace. We argue that the bottleneck lies not only in evidence discovery but also in its preservation and organization before answer generation. We propose SCoRE (Selection and Consolidation for Robust Evidence), a unified agent loop for explicit evidence selection and consolidation. During exploration, SCoRE retains only query-relevant observations and their source pointers in a maintained textual ledger, preserving earlier evidence while keeping the visual context bounded. At termination, it reloads the referenced original images and consolidates the visual evidence for answering, arranging it into a logical sequence. This decouples final reasoning from exploratory trial-and-error while ensuring strict visual grounding via indexed claim-to-image linkages. To enable end-to-end optimization of this unified rollout, our training paradigm combines filtered cold-start trajectory distillation with evidence-aware reinforcement learning, whose reward promotes evidence coverage, consolidation compactness, and answer correctness.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.15800 — 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 2026-10-15T22:29:03.362Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).