Metraipolis › archive › dc_8acbbf25de29

ReactionBench: A Human-Anchored Benchmark for Modeling Reactions to Video — arXiv:2611.00050

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.00050
License (as archived)
CC0-1.0
Archived
2026-11-03T10:24:27.177Z
SHA-256 of the archived text
10600cf81f6cb6025d76c3e33ce67ff7261f2a19a51d81480f03a48de55eb565
Stable URL of this page
https://metraipolis.com/documento/dc_8acbbf25de29

How to cite

ReactionBench: A Human-Anchored Benchmark for Modeling Reactions to Video — arXiv:2611.00050 — arXiv (en), https://arxiv.org/abs/2611.00050 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8acbbf25de29 (Metraipolis archive dc_8acbbf25de29, sha256 10600cf81f6cb6025d76c3e33ce67ff7261f2a19a51d81480f03a48de55eb565)
BibTeX
@misc{dc8acbbf25de29,
  title = {ReactionBench: A Human-Anchored Benchmark for Modeling Reactions to Video — arXiv:2611.00050},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.00050},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8acbbf25de29; sha256 10600cf81f6cb6025d76c3e33ce67ff7261f2a19a51d81480f03a48de55eb565},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00050. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Understanding what happens in a video is not sufficient for socially responsive multimodal systems: such systems also aim to relate what a person observes to how that person responds. Existing video benchmarks largely evaluate events, temporal reasoning, or people depicted in the video, while affective and viewer-response datasets often focus on emotion labels, aggregate responses, or prediction from the stimulus alone. This leaves an underexplored question: can current multimodal models ground an observed viewer reaction in the stimulus context associated with it? We introduce ReactionBench, a human-anchored benchmark for reaction-specific video understanding. Built from a candidate pool of 31,834 synchronized stimulus-reaction clips spanning 433 hours and eight domains, ReactionBench evaluates three complementary abilities: ranking the intensity of a viewer's reactions within a video, matching a reaction to its temporally aligned stimulus, and explaining the concrete stimulus cues linked to that reaction. Human annotation provides 14,894 ratings over 3,257 moments, and the frozen evaluation comprises 250 intensity profiles, 112 matching items, and 200 rationale items with 447 blind-written human references. Across seven multimodal models, reaction-specific grounding remains challenging: the best model reaches 50.9% accuracy on four-way stimulus-reaction matching, while GPT-5 shows no detectable loss under reaction substitution. Performance is also sensitive to evaluation protocol. Project page: this https URL

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.00050 — 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 2026-11-03T10:19:49.971Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).