# Decomposer: Learning to Decompile Symbolic Music to Programs

arXiv: https://arxiv.org/abs/2607.01849 · HTML: https://mlsanity.com/paper/2607.01849/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 64 of its 108 references that are in the map.
- First posted on arXiv 2 Jul 2026.

## Builds on (most cited first, 30 of 64)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Qwen2.5-Omni Technical Report](https://mlsanity.com/paper/2503.20215/) (2025)
- [Safe RLHF: Safe Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.12773/) (2023)
- [Texygen: A Benchmarking Platform for Text Generation Models](https://mlsanity.com/paper/1802.01886/) (2018)
- [CodeBLEU: a Method for Automatic Evaluation of Code Synthesis](https://mlsanity.com/paper/2009.10297/) (2020)
- [L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning](https://mlsanity.com/paper/2503.04697/) (2025)
- [Orca: Progressive Learning from Complex Explanation Traces of GPT-4](https://mlsanity.com/paper/2306.02707/) (2023)
- [Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains](https://mlsanity.com/paper/2507.17746/) (2025)
- [Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging](https://mlsanity.com/paper/2310.11564/) (2023)
- [Training Language Models to Reason Efficiently](https://mlsanity.com/paper/2502.04463/) (2025)
- [GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization](https://mlsanity.com/paper/2601.05242/) (2026)
- [DDSP: Differentiable Digital Signal Processing](https://mlsanity.com/paper/2001.04643/) (2020)
- [Checklists Are Better Than Reward Models For Aligning Language Models](https://mlsanity.com/paper/2507.18624/) (2025)
- [pix2code: Generating Code from a Graphical User Interface Screenshot](https://mlsanity.com/paper/1705.07962/) (2017)
- [CodeBERTScore: Evaluating Code Generation with Pretrained Models of Code](https://mlsanity.com/paper/2302.05527/) (2023)
- [LLM-Rubric: A Multidimensional, Calibrated Approach to Automated Evaluation of Natural Language Texts](https://mlsanity.com/paper/2501.00274/) (2024)
- [Rewarding the Unlikely: Lifting GRPO Beyond Distribution Sharpening](https://mlsanity.com/paper/2506.02355/) (2025)
- [Jointly Reinforcing Diversity and Quality in Language Model Generations](https://mlsanity.com/paper/2509.02534/) (2025)
- [Mind2Web 2: Evaluating Agentic Search with Agent-as-a-Judge](https://mlsanity.com/paper/2506.21506/) (2025)
- [Learning to Infer Graphics Programs from Hand-Drawn Images](https://mlsanity.com/paper/1707.09627/) (2017)
- [MT3: Multi-Task Multitrack Music Transcription](https://mlsanity.com/paper/2111.03017/) (2021)
- [CAD-Recode: Reverse Engineering CAD Code from Point Clouds](https://mlsanity.com/paper/2412.14042/) (2024)
- [NoveltyBench: Evaluating Language Models for Humanlike Diversity](https://mlsanity.com/paper/2504.05228/) (2025)
- [madmom: a new Python Audio and Music Signal Processing Library](https://mlsanity.com/paper/1605.07008/) (2016)
- [CAD-GPT: Synthesising CAD Construction Sequence with Spatial Reasoning-Enhanced Multimodal LLMs](https://mlsanity.com/paper/2412.19663/) (2024)
- [Just Enough Thinking: Efficient Reasoning with Adaptive Length Penalties Reinforcement Learning](https://mlsanity.com/paper/2506.05256/) (2025)
- [Base Models Beat Aligned Models at Randomness and Creativity](https://mlsanity.com/paper/2505.00047/) (2025)
- [LLM4Decompile: Decompiling Binary Code with Large Language Models](https://mlsanity.com/paper/2403.05286/) (2024)
- [Rendering-Aware Reinforcement Learning for Vector Graphics Generation](https://mlsanity.com/paper/2505.20793/) (2025)

## Abstract (arXiv)

Musical performance involves executing a set of high-level musical instructions, yet recovering those instructions from the performance is a challenging inverse problem. We present Decomposer, a post-training framework for symbolic music decompilation: the task of recovering executable, editable music programs from symbolic music. We instantiate the task as MIDI-to-Strudel decompilation, where the model takes symbolic MIDI as input and produces a program in Strudel, a music programming language, that reconstructs the input when executed. The task poses two challenges: Strudel is a low-resource language with little naturally paired MIDI-code data, and optimizing faithful reconstruction of MIDI alone can collapse to unreadable note-by-note transliteration. We address these challenges in two stages. First, we construct Strudel-Synth, a synthetic corpus of paired Strudel programs and rendered MIDI, and use it for supervised fine-tuning. Second, we refine the model with reinforcement learning on unpaired MIDI, optimizing rewards for both MIDI reconstruction faithfulness and code readability. Our evaluation across synthetic and real-world MIDI benchmarks shows that Decomposer achieves substantially higher MIDI reconstruction faithfulness than closed-source LLMs while producing more readable and diverse code than the heuristic converter.