ISMIR 2026 tutorial on LLMs × Music releases full book, slides and code
affige_yang · x · 2026-11-15
At ISMIR 2026 in Abu Dhabi, Seungheon Doh and colleagues presented the tutorial "LLMs × Music: Scalable Language Modeling Unlocks Music Applications", with all materials publicly released.
- Foundations: modern LLMs from large-scale pretraining to instruction tuning and preference alignment
- Cross-modal understanding: connecting audio, symbolic notation, score images and textual metadata
- Agentic pipelines: reasoning, tool calling and environment interaction for music tasks
- Applications: conversational recommendation for listeners, collaborative composition for creators, personalized assessment and coaching for learners
Code examples cover LLM-based music annotation, agentic music recommendation and a teaching agent — a solid entry point for MIR researchers and builders.
More from Research
- Epoch AI: post-training enhancements yield gains equal to 5-20x compute at under 1% the cost — CFGeek · 2026-11-15
- Qwen3 at 2.7 bits/weight with Leech lattice quantization beats IQ2_S on quality, loses on speed — Fuzzy-Layer9967 · 2026-11-15
- Steering with Natural-Scale Gaussian Noise in Random 4-Dim Space Sparks Follow-up Research — Sauers_ · 2026-11-15
- What 'Compute Multipliers' Mean: The Under-discussed Algorithmic Efficiency Metric in AI — CFGeek · 2026-11-15
- Researchers push back on Dwarkesh: defining AGI metrics is the hard part — dhadfieldmenell · 2026-11-15
- Meta FAIR's e-means beats k-means in 1/10th the epochs with 10-45x GPU speedups — TheTuringPost · 2026-11-15