Metraipolis › archive › dc_847754a39686

Recognition-Conditioned Reasoning: A Training-Free Multimodal-LLM Pipeline for Fine-Grained Micro-Action Understanding — arXiv:2608.21022

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.21022
License (as archived)
CC0-1.0
Archived
2026-10-18T10:56:18.868Z
SHA-256 of the archived text
f22d693bf79314448fb4ffa9b67ae79aa3a1492d4627e7e724230e8c76618b71
Stable URL of this page
https://metraipolis.com/documento/dc_847754a39686

How to cite

Recognition-Conditioned Reasoning: A Training-Free Multimodal-LLM Pipeline for Fine-Grained Micro-Action Understanding — arXiv:2608.21022 — arXiv (en), https://arxiv.org/abs/2608.21022 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_847754a39686 (Metraipolis archive dc_847754a39686, sha256 f22d693bf79314448fb4ffa9b67ae79aa3a1492d4627e7e724230e8c76618b71)
BibTeX
@misc{dc847754a39686,
  title = {Recognition-Conditioned Reasoning: A Training-Free Multimodal-LLM Pipeline for Fine-Grained Micro-Action Understanding — arXiv:2608.21022},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.21022},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_847754a39686; sha256 f22d693bf79314448fb4ffa9b67ae79aa3a1492d4627e7e724230e8c76618b71},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.21022. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Micro-actions are subtle, short, low-amplitude body movements, such as a fidgeting hand or a slight head tilt, that humans perform with little conscious intent yet that reliably leak emotional and psychological state. Understanding them goes beyond assigning a label: a model must also describe which body parts move and reason, faithfully, about why a clip warrants a particular fine-grained category. We present the training-free, prompt-only system that won first place in the fine-grained understanding track (MA-Bench) of the MAC~2026 Micro-Action Challenge, where both fine-tuning and ground-truth supervision are disallowed. Built entirely upon frozen multimodal large language models (MLLMs), the system dynamically routes each of the eight sub-tasks to the MLLM empirically best suited for that task: a discriminative MLLM for closed-ended recognition tasks and a generative MLLM for open-ended description and reasoning tasks. This architecture achieves a statistically significant performance advantage on open-ended tasks, attaining an average score of 2.68 (on a five-point scale) compared to 1.44 for the second-best approach.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-24 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Multimedia (cs.MM).
Pagina del lavoro: https://arxiv.org/abs/2608.21022 — 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 2026-10-18T10:48:13.589Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).