Metraipolis › archive › dc_becbc760418f

Multi-Agent Target-Existence Verification and Learned Mask Geometry Refinement: Winning Report of the MeViS-Text Track at the 8th LSVOS Challenge 2026 — arXiv:2608.11458

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.11458
License (as archived): CC0-1.0
Archived: 2026-10-19T05:49:19.772Z
SHA-256 of the archived text: 2c474f92b3e93b9fb0f77bb97a5e08121bca5886a65d67704390f2e01a5242f3
Stable URL of this page: https://metraipolis.com/documento/dc_becbc760418f

How to cite

Multi-Agent Target-Existence Verification and Learned Mask Geometry Refinement: Winning Report of the MeViS-Text Track at the 8th LSVOS Challenge 2026 — arXiv:2608.11458 — arXiv (en), https://arxiv.org/abs/2608.11458 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_becbc760418f (Metraipolis archive dc_becbc760418f, sha256 2c474f92b3e93b9fb0f77bb97a5e08121bca5886a65d67704390f2e01a5242f3)

BibTeX

@misc{dcbecbc760418f,
  title = {Multi-Agent Target-Existence Verification and Learned Mask Geometry Refinement: Winning Report of the MeViS-Text Track at the 8th LSVOS Challenge 2026 — arXiv:2608.11458},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.11458},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_becbc760418f; sha256 2c474f92b3e93b9fb0f77bb97a5e08121bca5886a65d67704390f2e01a5242f3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11458. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present the first-place solution to the MeViS-Text track of the 8th Large-scale Video Object Segmentation (LSVOS) Challenge 2026: referring video object segmentation guided by written motion expressions, including deceptive no-target expressions that match no object in the video and must yield empty masks in every frame. Our pipeline, SSUPER, resolves each expression into a visual concept, generates full-video candidate masklets with SAM~3.1, and selects target IDs. At every reasoning stage, three heterogeneous multimodal large language models independently execute the same stage-specific prompt before a single synthesis pass commits one schema-validated verdict. Although this system rejects every no-target expression in validation, the leaderboard reveals that a substantial share of test no-target cases still slips through. The reason is that hard negatives name a plausible object and fail only under the complete temporal predicate, so when selection and existence are decided together, a category-plausible masklet anchors the verdict. Hence, we decouple existence verification into an independent multi-agent audit of the full predicate (category, count, action, trajectory, event order, and semantic role) that distinguishes absence from temporary invisibility, discounts apparent motion caused by camera movement, and requires contradicting evidence rather than mere uncertainty for a no-target verdict. Without any new segmentation call, this audit recovers most of the residual no-target errors. A training-data-only StyleRefiner then aligns mask geometry with the annotation style of MeViSv2 while preserving every presence decision by construction, showing that once the semantics are fixed, part of the remaining error is stylistic rather than semantic. The complete system reaches a Final score of 0.9081339614 on the official challenge leaderboard.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-13 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.11458 — 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 2026-10-19T05:42:25.861Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).