Metraipolis › archive › dc_ce3f67ab9ea2
ProtoLIP: From Sentence-Level to Object-Level Evidence Disentanglement — arXiv:2609.16284
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.16284
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T18:32:48.809Z
- SHA-256 of the archived text
db63ea5acae6610068be2e3cab22f0b862a5f9e1eeac17431ca6e64f53312a1e- Stable URL of this page
https://metraipolis.com/documento/dc_ce3f67ab9ea2
How to cite
ProtoLIP: From Sentence-Level to Object-Level Evidence Disentanglement — arXiv:2609.16284 — arXiv (en), https://arxiv.org/abs/2609.16284 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ce3f67ab9ea2 (Metraipolis archive dc_ce3f67ab9ea2, sha256 db63ea5acae6610068be2e3cab22f0b862a5f9e1eeac17431ca6e64f53312a1e)
BibTeX
@misc{dcce3f67ab9ea2,
title = {ProtoLIP: From Sentence-Level to Object-Level Evidence Disentanglement — arXiv:2609.16284},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.16284},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ce3f67ab9ea2; sha256 db63ea5acae6610068be2e3cab22f0b862a5f9e1eeac17431ca6e64f53312a1e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.16284. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Query-conditioned vision-language models enable fine-grained interpretation by revealing which visual content supports a given textual query and how this evidence changes across queries. However, semantically, sentence-level evidence does not necessarily decompose into object-specific contributions, while spatially, object-level evidence can remain entangled with co-occurring objects and surrounding scene context. Across multiple VLM architectures and independent benchmarks, we observe persistent object-level evidence entanglement. Moreover, exposed evidence maps do not necessarily correspond to the evidence that directly constitutes the model's prediction. To disentangle visual evidence at both semantic and spatial levels, we introduce ProtoLIP, a lightweight prototype-mediated evidence layer that organizes reusable visual prototypes into text-derived semantic families and uses coarse-to-fine evidence routing, where semantic families constrain prototype eligibility and the complete query determines fine-grained prototype contributions. Our studies show that ProtoLIP improves evidence localization and separation across query granularities, achieving average relative gains of 29% in Pointing and 43% in Energy across four object- and phrase-level OOD benchmarks. Its localization gains also transfer to independently pretrained VLMs, with larger improvements observed in several transfer settings. On the primary backbone, ProtoLIP also improves image-text matching discrimination while remaining competitive with a spatially supervised grounding model in object-level localization. Crucially, ProtoLIP constructs its image-text matching score directly from localized prototype evidence, enabling exact decomposition across prototypes, semantic families, and spatial evidence without spatial annotations or backbone retraining. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-16 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.16284 — 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 2026-10-15T18:03:09.607Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).