Training-Efficient Text-to-Music Generation with State-Space Modeling

• URL: http://arxiv.org/abs/2601.14786v1
• Date: Wed, 21 Jan 2026 09:06:55 GMT
• Title: Training-Efficient Text-to-Music Generation with State-Space Modeling
• Authors: Wei-Jaw Lee, Fang-Chih Hsieh, Xuanjun Chen, Fang-Duo Tsai, Yi-Hsuan Yang,
• Abstract summary: We propose an open-source generative model backbone that is more training- and data-efficient.<n>We replace the Transformer backbone with the emerging class of state-space models (SSMs)<n>All proposed models are trained from scratch on a purely public dataset comprising 457 hours of CC-licensed music.
• Score: 8.067590027463497
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in text-to-music generation (TTM) have yielded high-quality results, but often at the cost of extensive compute and the use of large proprietary internal data. To improve the affordability and openness of TTM training, an open-source generative model backbone that is more training- and data-efficient is needed. In this paper, we constrain the number of trainable parameters in the generative model to match that of the MusicGen-small benchmark (with about 300M parameters), and replace its Transformer backbone with the emerging class of state-space models (SSMs). Specifically, we explore different SSM variants for sequence modeling, and compare a single-stage SSM-based design with a decomposable two-stage SSM/diffusion hybrid design. All proposed models are trained from scratch on a purely public dataset comprising 457 hours of CC-licensed music, ensuring full openness. Our experimental findings are three-fold. First, we show that SSMs exhibit superior training efficiency compared to the Transformer counterpart. Second, despite using only 9% of the FLOPs and 2% of the training data size compared to the MusicGen-small benchmark, our model achieves competitive performance in both objective metrics and subjective listening tests based on MusicCaps captions. Finally, our scaling-down experiment demonstrates that SSMs can maintain competitive performance relative to the Transformer baseline even at the same training budget (measured in iterations), when the model size is reduced to four times smaller. To facilitate the democratization of TTM research, the processed captions, model checkpoints, and source code are available on GitHub via the project page: https://lonian6.github.io/ssmttm/.

Related papers

• Exploring the Design Space of Transition Matching [28.72249421384186]
  We present a large-scale, systematic investigation into the design, training and sampling of the head in Transition Matching (TM) frameworks.<n>TM is an emerging paradigm for generative modeling that generalizes diffusion and flow-matching models as well as continuous-state autoregressive models.
  arXiv  Detail & Related papers  (2025-12-13T21:34:47Z)
• Satori-SWE: Evolutionary Test-Time Scaling for Sample-Efficient Software Engineering [51.7496756448709]
  Language models (LMs) perform well on coding benchmarks but struggle with real-world software engineering tasks.<n>Existing approaches rely on supervised fine-tuning with high-quality data, which is expensive to curate at scale.<n>We propose Test-Time Scaling (EvoScale), a sample-efficient method that treats generation as an evolutionary process.
  arXiv  Detail & Related papers  (2025-05-29T16:15:36Z)
• OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging [124.91183814854126]
  Model merging seeks to combine multiple expert models into a single model.<n>We introduce a benchmark for model merging research that clearly divides the tasks for MLLM training and evaluation.<n>We find that model merging offers a promising way for building improved MLLMs without requiring training data.
  arXiv  Detail & Related papers  (2025-05-26T12:23:14Z)
• Boosting Virtual Agent Learning and Reasoning: A Step-Wise, Multi-Dimensional, and Generalist Reward Model with Benchmark [72.46357004059661]
  Generalist Virtual Agents (GVAs) have shown significant promise in autonomous task execution.<n>To address these challenges, we propose Similar, a Step-Wise Multi-Dimensional Generalist Reward Model.<n>Similar offers fine-grained signals for agent training and can choose better action for inference-time scaling.
  arXiv  Detail & Related papers  (2025-03-24T13:30:47Z)
• Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling [191.7830199016589]
  We introduce InternVL 2.5, an advanced multimodal large language model (MLLM) series that builds upon InternVL 2.0.<n>InternVL 2.5 exhibits competitive performance, rivaling leading commercial models such as GPT-4o and Claude-3.5-Sonnet.<n>We hope this model contributes to the open-source community by setting new standards for developing and applying multimodal AI systems.
  arXiv  Detail & Related papers  (2024-12-06T18:57:08Z)
• LLäMmlein: Transparent, Compact and Competitive German-Only Language Models from Scratch [3.7160688974577156]
  We create two German-only decoder models, LL"aMmlein 120M and 1B, transparently from scratch and publish them, along with the training data, for the German NLP research community to use.<n>The model training involved several key steps, including extensive data preprocessing, the creation of a custom German tokenizer, the training itself, as well as the evaluation of the final models on various benchmarks.
  arXiv  Detail & Related papers  (2024-11-17T20:44:34Z)
• EMR-Merging: Tuning-Free High-Performance Model Merging [55.03509900949149]
  We show that Elect, Mask & Rescale-Merging (EMR-Merging) shows outstanding performance compared to existing merging methods. EMR-Merging is tuning-free, thus requiring no data availability or any additional training while showing impressive performance.
  arXiv  Detail & Related papers  (2024-05-23T05:25:45Z)
• Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series [11.635608108358575]
  We introduce Tiny Time Mixers (TTM), a compact model with effective transfer learning capabilities, trained exclusively on public TS datasets. TTM incorporates innovations like adaptive patching, diverse resolution sampling, and resolution prefix tuning to handle pre-training on varied dataset resolutions. It outperforms existing popular benchmarks in zero/few-shot forecasting by (4-40%), while reducing computational requirements significantly.
  arXiv  Detail & Related papers  (2024-01-08T15:21:21Z)
• The Languini Kitchen: Enabling Language Modelling Research at Different Scales of Compute [66.84421705029624]
  We introduce an experimental protocol that enables model comparisons based on equivalent compute, measured in accelerator hours. We pre-process an existing large, diverse, and high-quality dataset of books that surpasses existing academic benchmarks in quality, diversity, and document length. This work also provides two baseline models: a feed-forward model derived from the GPT-2 architecture and a recurrent model in the form of a novel LSTM with ten-fold throughput.
  arXiv  Detail & Related papers  (2023-09-20T10:31:17Z)
• Efficient GPT Model Pre-training using Tensor Train Matrix Representation [65.96485282393361]
  Large-scale transformer models feature billions of parameters, leading to difficulties in their deployment and prohibitive training costs from scratch. To reduce the number of parameters in the GPT-2 architecture, we replace the matrices of fully-connected layers with the corresponding Train Matrix(TTM) structure. The resulting GPT-based model stores up to 40% fewer parameters, showing the perplexity comparable to the original model.
  arXiv  Detail & Related papers  (2023-06-05T08:38:25Z)
• Imputing Knowledge Tracing Data with Subject-Based Training via LSTM Variational Autoencoders Frameworks [6.24828623162058]
  We adopt a subject-based training method to split and impute data by student IDs instead of row number splitting. We leverage two existing deep generative frameworks, namely variational Autoencoders (VAE) and Longitudinal Variational Autoencoders (LVAE) We demonstrate that the generated data from LSTM-VAE and LSTM-LVAE can boost the original model performance by about 50%.
  arXiv  Detail & Related papers  (2023-02-24T21:56:03Z)
• Online Model Compression for Federated Learning with Large Models [8.48327410170884]
  Online Model Compression (OMC) is a framework that stores model parameters in a compressed format and decompresses them only when needed. OMC can reduce memory usage and communication cost of model parameters by up to 59% while attaining comparable accuracy and training speed when compared with full-precision training.
  arXiv  Detail & Related papers  (2022-05-06T22:43:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.