Salvato in:
Dettagli Bibliografici
Autori principali: Bu, Fanjun, Tsai, Melina, Tjokro, Audrey, Bhattacharjee, Tapomayukh, Ortiz, Jorge, Ju, Wendy
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2512.07177
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911306940940288
author Bu, Fanjun
Tsai, Melina
Tjokro, Audrey
Bhattacharjee, Tapomayukh
Ortiz, Jorge
Ju, Wendy
author_facet Bu, Fanjun
Tsai, Melina
Tjokro, Audrey
Bhattacharjee, Tapomayukh
Ortiz, Jorge
Ju, Wendy
contents Robots operating in everyday environments must often decide when and whether to engage with people, yet such decisions often hinge on subtle nonverbal cues that unfold over time and are difficult to model explicitly. Drawing on a five-day Wizard-of-Oz deployment of a mobile service robot in a university cafe, we analyze how people signal interaction readiness through nonverbal behaviors and how expert wizards use these cues to guide engagement. Motivated by these observations, we propose a two-stage pipeline in which lightweight perceptual detectors (gaze shifts and proxemics) are used to selectively trigger heavier video-based vision-language model (VLM) queries at socially meaningful moments. We evaluate this pipeline on replayed field interactions and compare two prompting strategies. Our findings suggest that selectively using VLMs as proxies for social reasoning enables socially responsive robot behavior, allowing robots to act appropriately by attending to the cues people naturally provide in real-world interactions.
format Preprint
id arxiv_https___arxiv_org_abs_2512_07177
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Using Vision-Language Models as Proxies for Social Intelligence in Human-Robot Interaction
Bu, Fanjun
Tsai, Melina
Tjokro, Audrey
Bhattacharjee, Tapomayukh
Ortiz, Jorge
Ju, Wendy
Robotics
Robots operating in everyday environments must often decide when and whether to engage with people, yet such decisions often hinge on subtle nonverbal cues that unfold over time and are difficult to model explicitly. Drawing on a five-day Wizard-of-Oz deployment of a mobile service robot in a university cafe, we analyze how people signal interaction readiness through nonverbal behaviors and how expert wizards use these cues to guide engagement. Motivated by these observations, we propose a two-stage pipeline in which lightweight perceptual detectors (gaze shifts and proxemics) are used to selectively trigger heavier video-based vision-language model (VLM) queries at socially meaningful moments. We evaluate this pipeline on replayed field interactions and compare two prompting strategies. Our findings suggest that selectively using VLMs as proxies for social reasoning enables socially responsive robot behavior, allowing robots to act appropriately by attending to the cues people naturally provide in real-world interactions.
title Using Vision-Language Models as Proxies for Social Intelligence in Human-Robot Interaction
topic Robotics
url https://arxiv.org/abs/2512.07177