VIBE: Can a VLM Read the Room?

TL;DR

This paper investigates the ability of Vision Language Models to understand social cues and reasoning in social situations, highlighting a specific gap and proposing a new task and dataset to evaluate this capability.

Contribution

The paper introduces the Visual Social-Pragmatic Inference task and a high-quality dataset to assess VLMs' social reasoning abilities, addressing a previously overlooked limitation.

Findings

VLMs show limited performance on social reasoning tasks.

The Visual Social-Pragmatic Inference gap is a significant challenge.

Benchmark results reveal room for improvement in VLM social understanding.

Abstract

Understanding human social behavior such as recognizing emotions and the social dynamics causing them is an important and challenging problem. While LLMs have made remarkable advances, they are limited to the textual domain and cannot account for the major role that non-verbal cues play in understanding social situations. Vision Language Models (VLMs) can potentially account for this gap, however their ability to make correct inferences over such social cues has received little attention. In this paper, we explore the capabilities of VLMs at social reasoning. We identify a previously overlooked limitation in VLMs: the Visual Social-Pragmatic Inference gap. To target this gap, we propose a new task for VLMs: Visual Social-Pragmatic Inference. We construct a high quality dataset to test the abilities of a VLM for this task and benchmark the performance of several VLMs on it.