Metraipolis › archive › dc_953cb9560f32

Through a PRISM, Clearly: Step-Sequence Reasoning for Mistake Recognition in Procedural Videos — arXiv:2610.31117

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.31117
License (as archived)
CC0-1.0
Archived
2026-10-23T15:21:22.697Z
SHA-256 of the archived text
0f9ddc72fabd20c78c3c81fe1dc3107a7a101621df504a900bb68c3676c9d85b
Stable URL of this page
https://metraipolis.com/documento/dc_953cb9560f32

How to cite

Through a PRISM, Clearly: Step-Sequence Reasoning for Mistake Recognition in Procedural Videos — arXiv:2610.31117 — arXiv (en), https://arxiv.org/abs/2610.31117 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_953cb9560f32 (Metraipolis archive dc_953cb9560f32, sha256 0f9ddc72fabd20c78c3c81fe1dc3107a7a101621df504a900bb68c3676c9d85b)
BibTeX
@misc{dc953cb9560f32,
  title = {Through a PRISM, Clearly: Step-Sequence Reasoning for Mistake Recognition in Procedural Videos — arXiv:2610.31117},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.31117},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_953cb9560f32; sha256 0f9ddc72fabd20c78c3c81fe1dc3107a7a101621df504a900bb68c3676c9d85b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31117. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Procedural mistake recognition requires identifying deviations from valid multi-step activities. We study post-completion recognition from videos with benchmark-provided step boundaries. We introduce PRISM (Procedural mistake Recognition via Interpretable Semantic Modeling), which grounds video segments in language and maps their descriptions to procedural states using a lightweight hidden Markov model trained on correct executions. Reference alignment and semantic checks identify errors in step content and order. To support evaluation across heterogeneous benchmarks, we map dataset-specific annotations to four edit-based error types: insertion, deletion, substitution, and transposition. Across three benchmarks, we compare gold and VLM-generated descriptions in raw, semantic-role, and action-object forms. Decoder controls show consistent gains from the trained local decoder over nearest-candidate matching, with additional, dataset-dependent benefits from sequence decoding. Semantic role labeling improves Any-F1 in all six matched grounded comparisons, although the gains vary across datasets and do not translate into uniform improvements under separate configuration selection. These results establish language grounding as an effective interface for explicit procedural mistake reasoning and characterize the contributions of representation and decoding.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.31117 — 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 2026-10-23T15:09:38.393Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).