Towards Aligning Multimodal LLMs with Human Experts: A Focus on Parent-Child Interaction

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Shi, Weiyan, Choo, Kenny Tsu Wei
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912936378761216
author Shi, Weiyan
Choo, Kenny Tsu Wei
author_facet Shi, Weiyan
Choo, Kenny Tsu Wei
contents While multimodal large language models (MLLMs) are increasingly applied in human-centred AI systems, their ability to understand complex social interactions remains uncertain. We present an exploratory study on aligning MLLMs with speech-language pathologists (SLPs) in analysing joint attention in parent-child interactions, a key construct in early social-communicative development. Drawing on interviews and video annotations with three SLPs, we characterise how observational cues of gaze, action, and vocalisation inform their reasoning processes. We then test whether an MLLM can approximate this workflow through a two-stage prompting approach, separating observation from judgement. Our findings reveal that alignment is more robust at the observation layer, where experts share common descriptors, than at the judgement layer, where interpretive criteria diverge. We position this work as a case-based probe into expert-AI alignment in complex social behaviour, highlighting both the feasibility and the challenges of applying MLLMs to socially situated interaction analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2511_04366
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Aligning Multimodal LLMs with Human Experts: A Focus on Parent-Child Interaction
Shi, Weiyan
Choo, Kenny Tsu Wei
Human-Computer Interaction
Multimedia
While multimodal large language models (MLLMs) are increasingly applied in human-centred AI systems, their ability to understand complex social interactions remains uncertain. We present an exploratory study on aligning MLLMs with speech-language pathologists (SLPs) in analysing joint attention in parent-child interactions, a key construct in early social-communicative development. Drawing on interviews and video annotations with three SLPs, we characterise how observational cues of gaze, action, and vocalisation inform their reasoning processes. We then test whether an MLLM can approximate this workflow through a two-stage prompting approach, separating observation from judgement. Our findings reveal that alignment is more robust at the observation layer, where experts share common descriptors, than at the judgement layer, where interpretive criteria diverge. We position this work as a case-based probe into expert-AI alignment in complex social behaviour, highlighting both the feasibility and the challenges of applying MLLMs to socially situated interaction analysis.
title Towards Aligning Multimodal LLMs with Human Experts: A Focus on Parent-Child Interaction
topic Human-Computer Interaction
Multimedia
url https://arxiv.org/abs/2511.04366