Tag
CUE-Bench is a Chinese benchmark for affective stance that models explicit-implicit polarity interaction and provides intent and fine-grained emotion annotations, showing gains in emotion recognition and pragmatic intent detection.
This paper introduces ArtECulture, a benchmark for culture-conditioned visual emotion understanding in multimodal large language models, covering English, Chinese, and Arabic cultures with balanced Western and non-Western artwork. Evaluations reveal the task remains challenging, and the authors propose a retrieval-augmented framework to inject cultural knowledge into MLLMs.
Introduces CAREBench, a benchmark grounded in appraisal theory to evaluate LLMs' emotion understanding through cognitive appraisal reasoning, revealing that current models struggle with reasoning and positive emotion recognition despite matching humans on some downstream tasks.