Tears or Cheers? Benchmarking LLMs via Culturally Elicited Distinct Affective Responses

TL;DR

This paper introduces CEDAR, a new multimodal benchmark for evaluating how well large language models understand culturally influenced emotional responses, highlighting current models' limitations in capturing cultural affective nuances.

Contribution

The paper presents CEDAR, a novel benchmark built from culturally elicited affective responses, and a pipeline using LLMs and human evaluation to measure models' cultural affective understanding.

Findings

Models show a dissociation between language consistency and cultural alignment.

Current models struggle with culturally grounded affective understanding.

CEDAR includes 10,962 instances across multiple languages and emotions.

Abstract

Culture serves as a fundamental determinant of human affective processing and profoundly shapes how individuals perceive and interpret emotional stimuli. Despite this intrinsic link extant evaluations regarding cultural alignment within Large Language Models primarily prioritize declarative knowledge such as geographical facts or established societal customs. These benchmarks remain insufficient to capture the subjective interpretative variance inherent to diverse sociocultural lenses. To address this limitation, we introduce CEDAR, a multimodal benchmark constructed entirely from scenarios capturing Culturally \underline{\textsc{E}}licited \underline{\textsc{D}}istinct \underline{\textsc{A}}ffective \underline{\textsc{R}}esponses. To construct CEDAR, we implement a novel pipeline that leverages LLM-generated provisional labels to isolate instances yielding cross-cultural emotional distinctions, and subsequently derives reliable ground-truth annotations through rigorous human evaluation. The resulting benchmark comprises 10,962 instances across seven languages and 14 fine-grained emotion categories, with each language including 400 multimodal and 1,166 text-only samples. Comprehensive evaluations of 17 representative multilingual models reveal a dissociation between language consistency and cultural alignment, demonstrating that culturally grounded affective understanding remains a significant challenge for current models.