Metraipolis › archive › dc_9770da7e64df

WeAgent-MMSearch: Native Text-Vision Interaction for Multimodal Search Agents — arXiv:2608.28062

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.28062
License (as archived)
CC0-1.0
Archived
2026-10-17T10:47:26.212Z
SHA-256 of the archived text
19421e936a6fdee228843c3052edcd4f53ff3bcc449881ab48877b3bbdf58a82
Stable URL of this page
https://metraipolis.com/documento/dc_9770da7e64df

How to cite

WeAgent-MMSearch: Native Text-Vision Interaction for Multimodal Search Agents — arXiv:2608.28062 — arXiv (en), https://arxiv.org/abs/2608.28062 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9770da7e64df (Metraipolis archive dc_9770da7e64df, sha256 19421e936a6fdee228843c3052edcd4f53ff3bcc449881ab48877b3bbdf58a82)
BibTeX
@misc{dc9770da7e64df,
  title = {WeAgent-MMSearch: Native Text-Vision Interaction for Multimodal Search Agents — arXiv:2608.28062},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.28062},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9770da7e64df; sha256 19421e936a6fdee228843c3052edcd4f53ff3bcc449881ab48877b3bbdf58a82},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.28062. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multimodal search agents extend parametric knowledge with newly emerging and long-tail evidence from the open web. Yet many existing agentic search environments often expose retrieved evidence only as text and omit tool-returned images from subsequent context, reducing visually grounded trajectories to text-only reasoning. Long-horizon interaction also compounds tool-call, response-length, timeout, and budget failures, which can discard salvageable trajectories, waste rollout computation, and disturb policy updates. To address these issues, we introduce WeAgent-Harness, a multimodal agentic harness that supports native text-vision interaction and runtime recovery. Retrieved images receive persistent disk references, allowing the model to inspect, process, and cite them throughout the trajectory. Based on this harness, we develop WeAgent-MMSearch, an integrated system spanning data construction, agentic post-training, and multimodal rollout. For data construction, a strong MLLM uses WeAgent-Harness to discover, synthesize, and verify MMSearch-style tasks and collect expert trajectories. During post-training, our Failure-Aware GSPO (FA-GSPO) recovers salvageable abnormal rollouts and filters invalid ones to improve bounded multimodal planning and search. We also introduce VisTarget-Bench, a 150-task human-verified benchmark that pairs each question with a held-out target image, distinguishing image-retrieval failures from visual-perception failures. Evaluation on VisTarget-Bench and seven public benchmarks shows that agentic post-training improves the average score by 19.22 points, enabling our model to outperform similarly sized open-source models and rival models with roughly ten times its parameter count.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.28062 — 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 2026-10-17T10:33:42.388Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).