Language-Informed Synthesis of Rational Agent Models for Grounded Theory-of-Mind Reasoning On-The-Fly
This work addresses the challenge of real-world social inference for AI systems, particularly in novel situations, by providing a framework that combines language and vision, though it is incremental in building on existing multimodal and Bayesian methods.
The paper tackles the problem of multimodal social reasoning by integrating linguistic and visual inputs to draw context-specific social inferences, resulting in a model that outperforms state-of-the-art methods in capturing human judgments across various cognitive science tasks.
Drawing real world social inferences usually requires taking into account information from multiple modalities. Language is a particularly powerful source of information in social settings, especially in novel situations where language can provide both abstract information about the environment dynamics and concrete specifics about an agent that cannot be easily visually observed. In this paper, we propose Language-Informed Rational Agent Synthesis (LIRAS), a framework for drawing context-specific social inferences that integrate linguistic and visual inputs. LIRAS frames multimodal social reasoning as a process of constructing structured but situation-specific agent and environment representations - leveraging multimodal language models to parse language and visual inputs into unified symbolic representations, over which a Bayesian inverse planning engine can be run to produce granular probabilistic judgments. On a range of existing and new social reasoning tasks derived from cognitive science experiments, we find that our model (instantiated with a comparatively lightweight VLM) outperforms ablations and state-of-the-art models in capturing human judgments across all domains.