Metraipolis › archive › dc_79aeb18f18f3
Coupled Scene Graph Anticipation: When With-Constraint Recall Hides Structural Inconsistency — arXiv:2611.06738
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.06738
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-05T18:22:44.248Z
- SHA-256 of the archived text
f27423a57a3ebeef111c960747421f060d997add790fe1e928cdb161b65d3a11- Stable URL of this page
https://metraipolis.com/documento/dc_79aeb18f18f3
How to cite
Coupled Scene Graph Anticipation: When With-Constraint Recall Hides Structural Inconsistency — arXiv:2611.06738 — arXiv (en), https://arxiv.org/abs/2611.06738 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_79aeb18f18f3 (Metraipolis archive dc_79aeb18f18f3, sha256 f27423a57a3ebeef111c960747421f060d997add790fe1e928cdb161b65d3a11)
BibTeX
@misc{dc79aeb18f18f3,
title = {Coupled Scene Graph Anticipation: When With-Constraint Recall Hides Structural Inconsistency — arXiv:2611.06738},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.06738},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_79aeb18f18f3; sha256 f27423a57a3ebeef111c960747421f060d997add790fe1e928cdb161b65d3a11},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.06738. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Scene Graph Anticipation (SGA) predicts object relationships in future video frames. Continuous-time SGA models evolve each subject-object pair with an independent Neural ODE, so relationships sharing an entity cannot influence one another during integration, and nothing prevents mutually exclusive predictions, such as "sitting on" and "standing on", for the same pair. We introduce CoupledSGA, which links such relationships through a Relationship Interaction Graph with graph attention inside the ODE vector field, and adds a Structural Consistency Loss ($\mathcal{L}_{SC}$) based on a matrix of mutually exclusive predicates. Developing CoupledSGA revealed a limitation of the evaluation protocol. With-Constraint evaluation keeps one predicate per group for each subject-object pair, so its R@K and mR@K cannot capture the within-group violations targeted by $\mathcal{L}_{SC}$: we prove they are unchanged whenever each group's winning predicate and its score are held fixed. We further show that the gradient of $\mathcal{L}_{SC}$ depends on predicate scores rather than ground-truth labels: for two-predicate rules it amplifies the ordering the model already holds, and when all scores in a group are equal it suppresses predicates in more exclusion rules more strongly, explaining why the penalty can lower recall while improving consistency. On Action Genome, activating $\mathcal{L}_{SC}$ after two warm-up epochs reduces R@20 from 42.21% to 35.18% while lowering the No-Constraint contradiction rate from 45.92% to 4.07% (matched-budget R@20: 40.48% CoupledSGA vs. 44.78% SceneSayerODE). CoupledSGA is a testbed, not a state-of-the-art claim: With-Constraint recall captures the cost of enforcing structural consistency but not the improvement. We recommend reporting No-Constraint results alongside a contradiction rate for methods that constrain predicate co-activation.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.06738 — 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 2026-11-05T18:15:49.042Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).