This paper introduces TabPack, an efficient MLP ensemble for tabular data that samples and trains MLPs with different hyperparameters in parallel and selects ensemble members on-the-fly during trainin…

This paper investigates the impact of batch sampling strategies using text and audio embeddings on text-to-music generation under low-data conditions.

This paper introduces a mechanistic neuronal network model for multilayer learning, offering biological insights and an alternative to backpropagation.

The paper introduces the Decan metric, a novel, information-theoretic approach for measuring creative diversity in AI outputs, which successfully detects diversity loss across different model fine-tun…

This paper introduces PolSeT, a Polish psychoacoustic and Music Information Retrieval dataset with 1901 descriptors and 18 instrument sound ratings.

This paper reviews the limitations of Deep Learning models in EEG analysis for epilepsy diagnosis and proposes Kolmogorov-Arnold Networks (KANs) as a solution.

The paper introduces the Designed Vocalizations Dataset for AI-based voice conversion research on non-human vocalizations and effects, providing a standardized test set and benchmark results.

The paper introduces CoRP, a gradient-free operator that consolidates the benefits of ensemble-based post-training methods into a single, deployable model update, significantly improving performance w…

The paper proposes a novel Meta-Quantum Ensemble (MQE) framework, which fuses outputs from Quantum Support Vector Machines (QSVMs) and Quantum Neural Networks (QNNs) using a Random Forest meta-learner…

This paper proposes RPPNet, a two-stage deep learning architecture for music generation with variable structural boundaries, which automatically derives grouping of Rhythm-Pitch Primitive sequences fr…

The paper introduces Loopie, two Mixture-of-Experts models that outperform vanilla Transformer baselines in looped Transformers, with extensive ablation studies and a strong reasoning pipeline.

Maestro Order is a model-agnostic orchestration harness that turns unreliable models into reliable problem-solving systems by composing them with structural primitives and a budget-aware controller.

This paper investigates neural activity during five auditory conditions using EEG recordings from a 5-year-old participant, revealing condition-specific modulation of neural oscillatory activity and d…

The paper introduces Mpipe, a method for multimodal-aware heterogeneous parallel scheduling in large-scale multimodal language model training, achieving significant speedups on Ascend 910C NPU cluster…

This paper proposes an ensemble of two source separators, FTRNN and TF-Locoformer, trained with mixture invariant training (MixIT), and introduces mixture-constrained max pooling (MCM) to improve bird…