Socratis: Are large multimodal models emotionally aware?

TL;DR

Socratis introduces a new benchmark for multimodal emotion understanding, capturing diverse human reactions to images and captions, and evaluates the ability of large models to generate emotionally relevant explanations.

Contribution

The paper presents Socratis, a comprehensive dataset with multi-emotion annotations and reasons, and benchmarks multimodal models' capacity to generate human-like emotional explanations.

Findings

Humans prefer human-written reasons over machine-generated ones.

Current models struggle to generate emotionally accurate explanations.

Existing captioning metrics do not align with human preferences.

Abstract

Existing emotion prediction benchmarks contain coarse emotion labels which do not consider the diversity of emotions that an image and text can elicit in humans due to various reasons. Learning diverse reactions to multimodal content is important as intelligent machines take a central role in generating and delivering content to society. To address this gap, we propose Socratis, a societal reactions benchmark, where each image-caption (IC) pair is annotated with multiple emotions and the reasons for feeling them. Socratis contains 18K free-form reactions for 980 emotions on 2075 image-caption pairs from 5 widely-read news and image-caption (IC) datasets. We benchmark the capability of state-of-the-art multimodal large language models to generate the reasons for feeling an emotion given an IC pair. Based on a preliminary human study, we observe that humans prefer human-written reasons over 2 times more often than machine-generated ones. This shows our task is harder than standard generation tasks because it starkly contrasts recent findings where humans cannot tell apart machine vs human-written news articles, for instance. We further see that current captioning metrics based on large vision-language models also fail to correlate with human preferences. We hope that these findings and our benchmark will inspire further research on training emotionally aware models.