Evaluating point-light biological motion in multimodal large language models

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kadambi, Akila, Iacoboni, Marco, Aziz-Zadeh, Lisa, Narayanan, Srini
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908562907725824
author Kadambi, Akila
Iacoboni, Marco
Aziz-Zadeh, Lisa
Narayanan, Srini
author_facet Kadambi, Akila
Iacoboni, Marco
Aziz-Zadeh, Lisa
Narayanan, Srini
contents Humans can extract rich semantic information from minimal visual cues, as demonstrated by point-light displays (PLDs), which consist of sparse sets of dots localized to key joints of the human body. This ability emerges early in development and is largely attributed to human embodied experience. Since PLDs isolate body motion as the sole source of meaning, they represent key stimuli for testing the constraints of action understanding in these systems. Here we introduce ActPLD, the first benchmark to evaluate action processing in MLLMs from human PLDs. Tested models include state-of-the-art proprietary and open-source systems on single-actor and socially interacting PLDs. Our results reveal consistently low performance across models, introducing fundamental gaps in action and spatiotemporal understanding.
format Preprint
id arxiv_https___arxiv_org_abs_2509_23517
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Evaluating point-light biological motion in multimodal large language models
Kadambi, Akila
Iacoboni, Marco
Aziz-Zadeh, Lisa
Narayanan, Srini
Computer Vision and Pattern Recognition
Artificial Intelligence
Humans can extract rich semantic information from minimal visual cues, as demonstrated by point-light displays (PLDs), which consist of sparse sets of dots localized to key joints of the human body. This ability emerges early in development and is largely attributed to human embodied experience. Since PLDs isolate body motion as the sole source of meaning, they represent key stimuli for testing the constraints of action understanding in these systems. Here we introduce ActPLD, the first benchmark to evaluate action processing in MLLMs from human PLDs. Tested models include state-of-the-art proprietary and open-source systems on single-actor and socially interacting PLDs. Our results reveal consistently low performance across models, introducing fundamental gaps in action and spatiotemporal understanding.
title Evaluating point-light biological motion in multimodal large language models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2509.23517