Large language models improve music emotion prediction with in-context learning

Researchers have explored the use of in-context learning (ICL) with large language models to improve the prediction of music's emotional valence and arousal. While arousal is generally predictable from audio, valence is more subjective and context-dependent. The study found that ICL, particularly when using textual data like lyrics and audio captions, significantly boosted valence prediction accuracy. A multimodal approach incorporating raw audio also showed promise, suggesting ICL is an effective inference-time strategy for current language models in predicting music emotion. AI

IMPACT Demonstrates LLMs' potential for nuanced, subjective analysis beyond traditional tasks, opening doors for AI in creative fields.

RANK_REASON Research paper published on arXiv detailing a new method for predicting music emotion using LLMs. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• arXiv
• In-Context Learning
• Hugging Face
• Large Language Models
• MER60 dataset

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →