Leveraging LLM Embeddings for Cross Dataset Label Alignment and Zero Shot Music Emotion Prediction

TL;DR

This paper introduces a novel approach for music emotion recognition that uses LLM embeddings for label alignment across datasets and enables zero-shot prediction on unseen categories, improving generalization.

Contribution

The method leverages LLM embeddings for cross-dataset label alignment and introduces an alignment regularization for better adaptation to unseen data.

Findings

Effective zero-shot inference on new datasets.

Improved label alignment across multiple datasets.

Enhanced model generalization to unseen categories.

Abstract

In this work, we present a novel method for music emotion recognition that leverages Large Language Model (LLM) embeddings for label alignment across multiple datasets and zero-shot prediction on novel categories. First, we compute LLM embeddings for emotion labels and apply non-parametric clustering to group similar labels, across multiple datasets containing disjoint labels. We use these cluster centers to map music features (MERT) to the LLM embedding space. To further enhance the model, we introduce an alignment regularization that enables dissociation of MERT embeddings from different clusters. This further enhances the model's ability to better adaptation to unseen datasets. We demonstrate the effectiveness of our approach by performing zero-shot inference on a new dataset, showcasing its ability to generalize to unseen labels without additional training.