Tag
Research paper examining how large language models express social emotions compared to human cultural norms, finding systematic misalignment where LLMs show inconsistent patterns of engaging vs. disengaging emotion expressivity across cultural personas (European American and Latin American) compared to human responses.
Introduces FunAudioLLM, a model family that combines SenseVoice for multilingual ASR, emotion detection, and audio event detection with CosyVoice for natural speech generation, enabling enhanced voice interactions with LLMs through open-source models and code.