Skip to yearly menu bar Skip to main content

Workshop: Foundation Models for Decision Making

MMToM-QA: Multimodal Theory of Mind Question Answering

Chuanyang Jin · Yutong Wu · Jing Cao · Jiannan Xiang · Yen-Ling Kuo · Zhiting Hu · Tomer Ullman · Antonio Torralba · Josh Tenenbaum · Tianmin Shu

[ ] [ Project Page ]
presentation: Foundation Models for Decision Making
Fri 15 Dec 6:15 a.m. PST — 3:30 p.m. PST


Theory of Mind (ToM), the cognitive ability to understand people's minds, is an essential ingredient for developing machines with human-level social intelligence. Recent machine learning models, particularly large language models, seem to show some aspects of ToM understanding. However, existing ToM benchmarks use unimodal datasets -- either video or text. Human ToM, on the other hand, is more than video or text understanding. People can flexibly reason about another person's mind based on conceptual representations (e.g., goals, beliefs, plans) extracted from any available data, which can include visual cues, linguistic narratives, or both. To address this, we introduce a multimodal Theory of Mind question answering (MMToM-QA) benchmark. MMToM-QA comprehensively evaluates machine ToM both on multimodal data and on different kinds of unimodal data about a person's activity in a household environment. To engineer multimodal ToM capacity, we propose a novel method, BIP-ALM (Bayesian Inverse Planning Accelerated by Language Models). BIP-ALM extracts unified representations from multimodal data and utilizes language models for scalable Bayesian inverse planning. We conducted a systematic comparison of human performance, BIP-ALM, and state-of-the-art models, including GPT-4. The experiments demonstrate that large language models and large multimodal models still lack robust ToM capacity. BIP-ALM, on the other hand, shows promising results, by leveraging the power of both model-based mental inference and language models.

Chat is not available.