# MIDI-LLM: Improving Text-to-MIDI Music Generation via Adapting Large Language Models

arXiv: https://arxiv.org/abs/2511.03942 · HTML: https://mlsanity.com/paper/2511.03942/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 45 references that are in the map.
- First posted on arXiv 6 Nov 2025.

## Builds on (most cited first, 30 of 38)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [Don't Stop Pretraining: Adapt Language Models to Domains and Tasks](https://mlsanity.com/paper/2004.10964/) (2020)
- [Qwen2.5-Omni Technical Report](https://mlsanity.com/paper/2503.20215/) (2025)
- [Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation](https://mlsanity.com/paper/2211.06687/) (2022)
- [CNN Architectures for Large-Scale Audio Classification](https://mlsanity.com/paper/1609.09430/) (2016)
- [Simple and Controllable Music Generation](https://mlsanity.com/paper/2306.05284/) (2023)
- [Fr'echet Audio Distance: A Metric for Evaluating Music Enhancement Algorithms](https://mlsanity.com/paper/1812.08466/) (2018)
- [Music Transformer](https://mlsanity.com/paper/1809.04281/) (2018)
- [Fast Timing-Conditioned Latent Audio Diffusion](https://mlsanity.com/paper/2402.04825/) (2024)
- [Simple and Scalable Strategies to Continually Pre-train Large Language Models](https://mlsanity.com/paper/2403.08763/) (2024)
- [Compound Word Transformer: Learning to Compose Full-Song Music over Dynamic Directed Hypergraphs](https://mlsanity.com/paper/2101.02402/) (2021)
- [Music ControlNet: Multiple Time-varying Controls for Music Generation](https://mlsanity.com/paper/2311.07069/) (2023)
- [This Time with Feeling: Learning Expressive Musical Performance](https://mlsanity.com/paper/1808.03715/) (2018)
- [ChatMusician: Understanding and Generating Music Intrinsically with LLM](https://mlsanity.com/paper/2402.16153/) (2024)
- [The Jazz Transformer on the Front Line: Exploring the Shortcomings of AI-composed Music through Quantitative Measures](https://mlsanity.com/paper/2008.01307/) (2020)
- [MuseCoco: Generating Symbolic Music from Text](https://mlsanity.com/paper/2306.00110/) (2023)
- [MuseMorphose: Full-Song and Fine-Grained Piano Music Style Transfer with One Transformer VAE](https://mlsanity.com/paper/2105.04090/) (2021)
- [Anticipatory Music Transformer](https://mlsanity.com/paper/2306.08620/) (2023)
- [MuPT: A Generative Symbolic Music Pretrained Transformer](https://mlsanity.com/paper/2404.06393/) (2024)
- [MusicMagus: Zero-Shot Text-to-Music Editing via Diffusion Models](https://mlsanity.com/paper/2402.06178/) (2024)
- [NotaGen: Advancing Musicality in Symbolic Music Generation with Large Language Model Training Paradigms](https://mlsanity.com/paper/2502.18008/) (2025)
- [Melody transcription via generative pre-training](https://mlsanity.com/paper/2212.01884/) (2022)
- [MidiCaps: A large-scale MIDI dataset with text captions](https://mlsanity.com/paper/2406.02255/) (2024)

## Built on by (most cited first, 1 of 1)

- [Libretto: Giving LLM Agents a Sense of Musical Structure](https://mlsanity.com/paper/2606.22708/) (2026)

## Abstract (arXiv)

We present MIDI-LLM, a recipe that improves multitrack text-to-MIDI generation via adapting Large Language Models (LLMs). MIDI-LLM expands an LLM's text vocabulary to include MIDI tokens and employs a two-stage training pipeline: (i) unimodal continued pretraining on music-adjacent text and standalone MIDIs, and (ii) multimodal supervised finetuning on text-MIDI pairs. Our instantiation of MIDI-LLM based on Llama 3.2 (1B) outperforms the recent Text2midi model in both text control and musical quality, and readily integrates with optimized inference ecosystems like vLLM. To align with real-world songwriting workflows, we further finetune our MIDI-LLM on the TheoryTab dataset for text-conditioned lead sheet (i.e., melody + chords) generation and infilling. A comprehensive ablation study validates the synergy between LLM text pretraining, standalone MIDI pretraining, and supervised text-to-MIDI finetuning. Finally, an in-the-wild blind user study conducted in a real-world creative workflow at scale with 58 Hookpad Aria users and 4,002 generated outputs demonstrates that our MIDI-LLM achieves the highest acceptance rate in zero-to-one lead sheet generation over baselines without text control or LLM pretraining, confirming its efficacy in human-AI music co-creation.