The paper introduces MI-MIDI 2, an end‑to‑end protocol for discovering and evaluating sparse autoencoder (SAE) features in text‑to‑MIDI generation. It applies the protocol to the text2midi and MIDI‑LLM models, characterizing SAE features through correlations with musical metrics, instrument programs, and automated descriptions, and then testing them via direct intervention in generated MIDI. The study finds that certain SAE features—particularly those linked to instrument programs and note‑density adjustments—effectively steer the output, with evidence validated across multiple controls and independent prompts.
arXiv:2608. 06638v1 Announce Type: cross Abstract: Mechanistic interpretability of music generation has concentrated on audio models, leaving symbolic models largely unexplored.
arXiv:2608. 04378v1 Announce Type: cross Abstract: Collaborative music agents need internal representations rich enough to support both understanding and generation, yet flexible enough for a workflow where the human retains agency.
SpanSynth-Edit is a flow‑matching model that enables MIDI‑guided synthesis and editing of multi‑instrument audio mixtures using low‑frame‑rate scalar‑quantised latents. It encodes instrument‑labelled note lifecycles as unordered event sets, pools them into a conditioning vector per audio‑latent frame, and uses contextual audio for instrument‑specific timbre guidance. The model supports editing by resynthesising target regions from revised MIDI and demonstrates competitive performance on single‑ and multi‑instrument benchmarks, including within‑frame onset control.
MIDI-REPA is a framework that aligns symbolic music representations with intermediate layers of a flow-matching DiT model to improve text-to-music generation. By deriving MIDI supervision from automatic music transcription and aligning a MuScriptor-derived piano roll with an early DiT layer throughout the flow trajectory, the method achieves competitive performance on the ATTM objective metrics without extra training data or techniques. The approach is trainable on a single RTX 3090 in about 31 hours and is supported by released transcription pipeline, training code, and audio samples.
arXiv:2606. 03169v1 Announce Type: cross Abstract: Recent song generation systems can synthesize realistic audio, yet generating complete songs remains challenging for two reasons.
The paper introduces CPR, a piano rendering framework that combines continuous autoregressive modeling with local flow matching and full‑sequence refinement. It predicts continuous hidden states, generates 24 kHz acoustic latents, and upsamples to 48 kHz, while new techniques BREPA and MT‑RoPE enhance musical semantics and cross‑modal alignment.
Neural music editing requires a system to perform a requested change while preserving the musical content that should remain unchanged. Existing evaluations often lack a paired target edit, making the...
DuET is a Dual-Encoder Transformer that jointly learns note-level piano performance-to-score transcription and frame-level beat/downbeat tracking. It uses bidirectional cross‑attention between note and frame encoders and an autoregressive decoder that alternates between their representations to predict structured score attributes. The model introduces TimeRoPE and Time‑Pitch Rotary Positional Encoding (TPRoPE) to share a performance‑time reference and encode relative temporal and pitch relationships, achieving significant error reductions on ASAP and Gangqinpu datasets.
arXiv:2512. 02652v2 Announce Type: replace-cross Abstract: Existing methods for expressive music performance rendering, a conditional generation task that aims to generate a human-like performance from a symbolic score, rely on supervised learning over small labeled datasets, which limits scaling of both data volume and model size, despite the availability of vast unlabeled music, as in vision and language.
arXiv:2607. 01849v1 Announce Type: cross Abstract: Musical performance involves executing a set of high-level musical instructions, yet recovering those instructions from the performance is a challenging inverse problem.
Piano2Pop is a music‑expansion task that converts piano MIDI into full instrumental pop productions. The method uses a pseudo‑supervised approach that conditions a pretrained flow‑matching model on piano rolls, trained on synthetic piano‑instrumental pairs derived from pop recordings. Experiments on 476 real‑world piano‑cover/pop pairs show that a pair‑supervised piano‑cover generator outperforms zero‑shot baselines, while function‑selective transcription (FST) reduces piano‑heavy outputs in pair‑free training and remains a promising alternative.
arXiv:2608. 09035v1 Announce Type: cross Abstract: Text-to-music generation has advanced rapidly, but current systems still rely primarily on global text prompts, leaving the structural organization of generated music implicit and difficult to inspect, control, or revise before audio generation.