Prof Chua Tat Seng Receives Best Paper Award at MMM (Multimedia Modeling) 2026

We received the Best Paper Award at MMM (Multimedia Modeling) 2026 held in Prague, Czech, on 28-31 January 2026. The detail of the paper is:

Integrating Symbolic and Waveform Music into Large Language Models, Tu Teng; Liu Xiaohao; Ma Yunshan; Qi Ji; and Chua Tat-Seng

This paper proposes a unified music language model named UniMuLM that integrates symbolic, waveform music, and textual instructions, and design a bar-level tokenizer to explore the fine-grained correlations between different modalities. It addresses the challenges of missing modalities and unstable training to adapt to different scenarios. The model has been found to exhibit superior performance compared to SOTA models across five musical tasks, over nine benchmark datasets.