Metraipolis › archive › dc_641e2d2258a5
Rethinking Multi-Image Re-Representation in Multi-Image Understanding — arXiv:2609.39363
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.39363
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T12:45:17.946Z
- SHA-256 of the archived text
81e71d950fa8cc33b4b64b2f7632688a27c54febc3d774e259b9564ce47378c5- Stable URL of this page
https://metraipolis.com/documento/dc_641e2d2258a5
How to cite
Rethinking Multi-Image Re-Representation in Multi-Image Understanding — arXiv:2609.39363 — arXiv (en), https://arxiv.org/abs/2609.39363 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_641e2d2258a5 (Metraipolis archive dc_641e2d2258a5, sha256 81e71d950fa8cc33b4b64b2f7632688a27c54febc3d774e259b9564ce47378c5)
BibTeX
@misc{dc641e2d2258a5,
title = {Rethinking Multi-Image Re-Representation in Multi-Image Understanding — arXiv:2609.39363},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.39363},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_641e2d2258a5; sha256 81e71d950fa8cc33b4b64b2f7632688a27c54febc3d774e259b9564ce47378c5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39363. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multi-image understanding requires MLLMs not only to recognise the content of individual images, but also to organise visual evidence distributed across them. We study this problem through multi-image re-representation, viewing prompted Chain-of-Thought reasoning and agentic visual tool use as different ways of re-organising visual evidence during reasoning. We introduce Mosaic, a general-purpose multi-image visual harness that enables an MLLM to actively construct visual intermediates with ten composable image operations. We compare five re-representation settings on existing multi-image benchmarks and on MosaicBench, a new grounding-focused benchmark for fine-grained multi-image understanding. Our experiments show that the relative benefits of textual and visual re-representation are strongly task-dependent. Visual re-representation is particularly effective for tasks requiring precise visual evidence, including hypothesis testing, precision comparison, and orientation-sensitive reasoning, while tasks dominated by higher-level semantic content show smaller or less consistent gains. Building on this finding, we train MosaicAgent-8B to use Mosaic with reinforcement learning using only accuracy and format rewards. Without demonstration trajectories or rewards for specific tool-use, the agent learns to compose visual operations over multiple steps and exhibits diverse problem-solving patterns unpromptedly. Code and data will be released at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.39363 — 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 2026-10-07T12:14:14.067Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).