Introduces a new lyrics dataset and hybrid human-LLM framework for emotion annotation that predicts misalignment to improve efficiency.
MERGE–a bimodal dataset for static music emotion recognition,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Feedback-driven alignment with numerical rewards improves MusicLLM emotion regression on arousal and valence over instruction tuning alone while preserving MusicQA performance.
citing papers explorer
-
A Hybrid Framework for Song Lyric Annotation Based on Human-LLM Alignment
Introduces a new lyrics dataset and hybrid human-LLM framework for emotion annotation that predicts misalignment to improve efficiency.