Metraipolis › archive › dc_a667b4eda93c
dKFD: Phase-Structured Evidence Allocation for Fixed-Budget Localized Event Understanding — arXiv:2609.33083
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.33083
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T23:51:16.370Z
- SHA-256 of the archived text
607c6e8bb4e5feee6d00b5fc0b7508ac59c064999620a203f7aceca6f59ddce4- Stable URL of this page
https://metraipolis.com/documento/dc_a667b4eda93c
How to cite
dKFD: Phase-Structured Evidence Allocation for Fixed-Budget Localized Event Understanding — arXiv:2609.33083 — arXiv (en), https://arxiv.org/abs/2609.33083 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a667b4eda93c (Metraipolis archive dc_a667b4eda93c, sha256 607c6e8bb4e5feee6d00b5fc0b7508ac59c064999620a203f7aceca6f59ddce4)
BibTeX
@misc{dca667b4eda93c,
title = {dKFD: Phase-Structured Evidence Allocation for Fixed-Budget Localized Event Understanding — arXiv:2609.33083},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.33083},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a667b4eda93c; sha256 607c6e8bb4e5feee6d00b5fc0b7508ac59c064999620a203f7aceca6f59ddce4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33083. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Sparse video understanding often requires selecting a small set of visual evidence under a fixed frame budget. Most sparse selectors allocate this budget globally, allowing all frames to compete with one another. For temporally localized events, this can be a poor inductive bias: useful evidence is often distributed across pre-event context, the event itself, and post-event consequences. We study fixed-budget evidence allocation for localized event videos and show that globally competitive Top-$K$ selectors can preserve recognition and grounding while producing unstable event evidence. On DoTA Video Anomaly Recognition, Global Top-$K$ obtains competitive recognition and temporal grounding, but low selector-event alignment at $K=12$ (Frame AUC $51.5 \pm 10.8$). We propose dKFD, a phase-structured differentiable selector that reserves evidence capacity across pre-event, event, and post-event phases after full-sequence temporal encoding. Under matched-budget multi-seed evaluation, dKFD improves Frame AUC by $+30.97$ over a matched Global Top-$K$ selector at $K=12$ ($p<0.01$), while yielding modest but statistically significant recognition gains and comparable temporal grounding. Mechanism ablations show that phase supervision is load-bearing: removing it reduces Frame AUC to $41.1 \pm 12.0$ even when phase-partitioned budgets are retained. Downstream diagnostics on VRU-Accident show consistent gains over learned Global Top-$K$ across VLM families, while dense captioning reveals a boundary condition where uniform sampling remains competitive. These results support phase-structured allocation as a controlled fixed-budget approach for event-centric sparse evidence selection, not as a universal video summarization strategy. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.33083 — 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 2026-10-07T23:29:27.136Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).