Metraipolis › archive › dc_62c58fb8a361

Compositional Benchmark Synthesis for Hierarchical Human Action Recognition — arXiv:2608.10765

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.10765
License (as archived)
CC0-1.0
Archived
2026-10-19T06:08:35.365Z
SHA-256 of the archived text
a3dc55049efddd34e62667f28f5baad2b4dfe2831e090a76d9a3a8efff28a4ee
Stable URL of this page
https://metraipolis.com/documento/dc_62c58fb8a361

How to cite

Compositional Benchmark Synthesis for Hierarchical Human Action Recognition — arXiv:2608.10765 — arXiv (en), https://arxiv.org/abs/2608.10765 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_62c58fb8a361 (Metraipolis archive dc_62c58fb8a361, sha256 a3dc55049efddd34e62667f28f5baad2b4dfe2831e090a76d9a3a8efff28a4ee)
BibTeX
@misc{dc62c58fb8a361,
  title = {Compositional Benchmark Synthesis for Hierarchical Human Action Recognition — arXiv:2608.10765},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.10765},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_62c58fb8a361; sha256 a3dc55049efddd34e62667f28f5baad2b4dfe2831e090a76d9a3a8efff28a4ee},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10765. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recognizing human behavior across levels of abstraction, from atomic actions to long-horizon intentions, requires data annotated along a semantic hierarchy. Large corpora provide isolated, atomically labeled clips without temporal composition, whereas recorded composite-activity corpora offer shallow, domain-narrow, fixedhierarchies. A benchmark-generation and evaluation frameworkis proposed that synthesizes a four-level hierarchical-intention benchmark, spanning actions, activities, low-level intentions (LLIs), and high-level intentions (HLIs), from a flat single-label action corpus while retaining real pre-extracted features at the action level. Episodes are assembled by a transition model under a subject-consistency constraint, and a coverage-aware sampler reduces the subject usage Gini from 0.566 to 0.248. Synthesizing such a benchmark raises a circular-supervision risk that recorded datasets avoid: if the rules generating the episodes also govern the evaluation, models can succeed by recovering the generator rather than through genuine reasoning. Validity is addressed by design, holding sequence-generation rules disjoint from the first-order-logic rules used at evaluation. The instantiation yields 15,002 episodes. Four reference baselines from different model families characterize difficulty, not as recognition methods. A compositional held-out gap of 0.13 to 0.17 macro-F1 appears across all baselines, including a graph-aware model that recognizes best yet does not close the gap, indicating a structural property of the benchmark rather than a model artifact. A logic-free baseline still violates the held-out semantic rules above their intrinsic data rate, and the order-destroying control changes macro-F1 within seed variation, serving as a generator-consistency check. Theontology, transition model, and generator are released so the benchmark can beregenerated and extended.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-12 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.10765 — 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 2026-10-19T05:57:28.294Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).