Natural language information needs over symbolic music scores rarely reduce to a single-step lookup. Many queries require compositional Music Information Retrieval (MIR) that extracts multiple pieces of evidence from structured notation and aggregates them to answer the question. This setting remains challenging for Large Language Models due to the mismatch between natural language intents and symbolic representations, as well as the difficulty of reliably handling long structured contexts. Existing benchmarks only partially capture these retrieval demands, often emphasizing isolated theoretical knowledge or simplified settings. We introduce CSyMR-Bench, a benchmark for compositional MIR in symbolic music reasoning grounded in authentic user scenarios. It contains 126 multiple-choice questions curated from community discussions and professional examinations, where each item requires chaining multiple atomic analyses over a score to derive implicit musical evidence. To support diagnosis, we provide a taxonomy with six query intent categories and six analytical dimension tags. We further propose a tool-augmented retrieval and reasoning framework, CSyMR-Agent, that integrates a ReAct-style controller with deterministic symbolic analysis operators built with music21. Experiments across prompting baselines and agent variants show that tool-grounded compositional retrieval consistently outperforms Large Language Model-only approaches, yielding 5-7% absolute accuracy gains, with the largest improvements on analysis-heavy categories.

Recent advances in Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities across various visionlanguage tasks.However, their reasoning abilities in the multimodal symbolic music domain remain largely unexplored.We introduce Wild-Score, the first in-the-wild multimodal symbolic music reasoning and analysis benchmark, designed to evaluate MLLMs' capacity to interpret real-world music scores and answer complex musicological queries.Each instance in WildScore is sourced from genuine musical compositions and accompanied by authentic user-generated questions and discussions, capturing the intricacies of practical music analysis.To facilitate a comprehensive evaluation, we propose a systematic taxonomy, comprising both high-level and fine-grained musicological ontologies.Furthermore, we frame complex music reasoning as multiple-choice question answering, enabling controlled and scalable assessment of MLLMs' symbolic music understanding.Empirical benchmarking of state-ofthe-art MLLMs on WildScore reveals intriguing patterns in their visual-symbolic reasoning, uncovering both promising directions and persistent challenges for MLLMs in symbolic music reasoning and analysis.We release the dataset 1 and code 2 .

The recent explosion of generative AI-Music systems has raised numerous concerns over data copyright, licensing music from musicians, and the conflict between open-source AI and large prestige companies. Such issues highlight the need for publicly available, copyright-free musical data, in which there is a large shortage, particularly for symbolic music data. To alleviate this issue, we present PDMX: a large-scale open-source dataset of over 250K public domain MusicXML scores collected from the score-sharing forum MuseScore, making it the largest available copyright-free symbolic music dataset to our knowledge. PDMX additionally includes a wealth of both tag and user interaction metadata, allowing us to efficiently analyze the dataset and filter for high quality user-generated scores. Given the additional metadata afforded by our data collection process, we conduct multitrack music generation experiments evaluating how different representative subsets of PDMX lead to different behaviors in downstream models, and how user-rating statistics can be used as an effective measure of data quality.

Modeling temporal characteristics plays a significant role in the representation learning of audio waveform. We propose Contrastive Long-form Language-Audio Pretraining (CoLLAP) to significantly extend the perception window for both the input audio (up to 5 minutes) and the language descriptions (exceeding 250 words), while enabling contrastive learning across modalities and temporal dynamics. Leveraging recent Music-LLMs to generate long-form music captions for full-length songs, augmented with musical temporal structures, we collect 51.3K audio-text pairs derived from the large-scale AudioSet training dataset, where the average audio length reaches 288 seconds. We propose a novel contrastive learning architecture that fuses language representations with structured audio representations by segmenting each song into clips and extracting their embeddings. With an attention mechanism, we capture multimodal temporal correlations, allowing the model to automatically weigh and enhance the final fusion score for improved contrastive alignment. Finally, we develop two variants of the CoLLAP model with different types of backbone language models. Through comprehensive experiments on multiple long-form music-text retrieval datasets, we demonstrate consistent performance improvement in retrieval accuracy compared with baselines. We also show the pretrained CoLLAP models can be transferred to various music information retrieval tasks, with heterogeneous long-form multimodal contexts.

Recent music large language models (music LLMs) have shown great potential in music understanding through large-scale multimodal pre-training. While some existing music LLMs have been augmented with temporally-aware music captions, music information retrieval (MIR) features conventionally do not exist in music caption datasets, thus neglected by music-LLMs. To bridge the gap between recent music LLMs and conventional music information retrieval tasks, we propose FUTGA-MIR (Fine-grained Music Understanding through Temporally-enhanced Generative Augmentation with Music Information Retrieval) to enhance the existing music LLMs by augmenting them with MIR features and aligning with human feedback. We calibrate the original positional distribution of music clips in the created pre-training synthetic music captions, conditioned on MIR features. The redistributed synthetic dataset better respects realistic MIR distributions of music and is tightly aligned with realistic music captions as well. In addition, We incorporate human-annotated music captions and MIR features from the Harmonixset dataset for an additional fine-tuning step, which addresses sim-to-real domain gaps and improves performance on realistic music tracks. We further evaluate FUTGA-MIR on several downstream tasks, including music classification, retrieval, and generation, and demonstrate the versatile capacities of FUTGA-MIR and better generation quality compared with previous music captioning models. Generated temporal-aware music descriptions are illustrated in our demonstration https://namburiamit.github.io/futga-music.github.io/.

Despite advances in diffusion-based text-to-music (TTM) methods, efficient, high-quality generation remains a challenge. We introduce Presto!, an approach to inference acceleration for score-based diffusion transformers via reducing both sampling steps and cost per step. To reduce steps, we develop a new score-based distribution matching distillation (DMD) method for the EDM-family of diffusion models, the first GAN-based distillation method for TTM. To reduce the cost per step, we develop a simple, but powerful improvement to a recent layer distillation method that improves learning via better preserving hidden state variance. Finally, we combine our step and layer distillation methods together for a dual-faceted approach. We evaluate our step and layer distillation methods independently and show each yield best-in-class performance. Our combined distillation method can generate high-quality outputs with improved diversity, accelerating our base model by 10-18x (230/435ms latency for 32 second mono/stereo 44.1kHz, 15x faster than the comparable SOTA model) — the fastest TTM to our knowledge.

Western music is an innately hierarchical system of interacting levels of structure, from fine-grained melody to high-level form. In order to analyze music compositions holistically and at multiple granularities, we propose a unified, hierarchical meta-representation of musical structure called the structural temporal graph (STG). For a single piece, the STG is a data structure that defines a hierarchy of progressively finer structural musical features and the temporal relationships between them. We use the STG to enable a novel approach for deriving a representative structural summary of a music corpus, which we formalize as a dually NP-hard combinatorial optimization problem. Our approach first applies simulated annealing to develop a measure of structural distance between two music pieces rooted in graph isomorphism. Our approach then combines the formal guarantees of SMT solvers with nested simulated annealing over structural distances to produce a structurally sound, representative centroid STG for an entire corpus of STGs from individual pieces. To evaluate our approach, we conduct experiments verifying that structural distance accurately differentiates between music pieces, and that derived centroids accurately structurally characterize their corpora.

Recent monocular metric depth estimation (MMDE) methods have made notable progress towards zero-shot generalization. However, they still exhibit a significant performance drop on out-of-distribution datasets. We address this limitation by injecting defocus blur cues at inference time into Marigold, a \textit{pre-trained} diffusion model for zero-shot, scale-invariant monocular depth estimation (MDE). Our method effectively turns Marigold into a metric depth predictor in a training-free manner. To incorporate defocus cues, we capture two images with a small and a large aperture from the same viewpoint. To recover metric depth, we then optimize the metric depth scaling parameters and the noise latents of Marigold at inference time using gradients from a loss function based on the defocus-blur image formation model. We compare our method against existing state-of-the-art zero-shot MMDE methods on a self-collected real dataset, showing quantitative and qualitative improvements.

We propose Diffusion Inference-Time T-Optimization (DITTO), a general-purpose framework for controlling pre-trained text-to-music diffusion models at inference-time via optimizing initial noise latents. Our method can be used to optimize through any differentiable feature matching loss to achieve a target (stylized) output and leverages gradient checkpointing for memory efficiency. We demonstrate a surprisingly wide-range of applications for music generation including inpainting, outpainting, and looping as well as intensity, melody, and musical structure control – all without ever fine-tuning the underlying model. When we compare our approach against related training, guidance, and optimization-based methods, we find DITTO achieves state-of-the-art performance on nearly all tasks, including outperforming comparable approaches on controllability, audio quality, and computational efficiency, thus opening the door for high-quality, flexible, training-free control of diffusion models. Sound examples can be found at https://ditto-music.github.io/web/.

Open vocabulary models (e.g. CLIP) have shown strong performance on zero-shot classification through their ability generate embeddings for each class based on their (natural language) names. Prior work has focused on improving the accuracy of these models through prompt engineering or by incorporating a small amount of labeled downstream data (via finetuning). However, there has been little focus on improving the richness of the class names themselves, which can pose issues when class labels are coarsely-defined and are uninformative. We propose Classification with Hierarchical Label Sets (or CHiLS), an alternative strategy for zero-shot classification specifically designed for datasets with implicit semantic hierarchies. CHiLS proceeds in three steps: (i) for each class, produce a set of subclasses, using either existing label hierarchies or by querying GPT-3; (ii) perform the standard zero-shot CLIP procedure as though these subclasses were the labels of interest; (iii) map the predicted subclass back to its parent to produce the final prediction. Across numerous datasets with underlying hierarchical structure, CHiLS leads to improved accuracy in situations both with and without ground-truth hierarchical information. CHiLS is simple to implement within existing zero-shot pipelines and requires no additional training cost. Code is available at: https://github.com/acmi-lab/CHILS.