Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
Researchers propose DSTFView, a dual-input spatio-temporal-frequency multi-view workload forecasting framework for collaborative cloud-edge environments, which jointly models closeness and period dependencies and extract…
A new multi-agent lane change decision model using the QMIX framework with a convolutional neural network (CNN-QMIX) can increase cooperative platooning rates by up to 26.2% for connected automated vehicles (CAVs) in mix…
Researchers propose CHiPS, a lightweight character-level authorship attribution method for Romanian texts that uses character histograms and positional signals without tokenization or pretrained language models. On a loc…
A new preprint on arXiv (2607.22748v1) introduces Accessibility Plasticity, a principle of adaptive computation in which systems adapt by reorganizing which existing computations can interact, rather than only modifying …
Researchers introduce MegaSlide-DiT, a prototype that adapts a pre-trained 105-billion-parameter Diffusion Transformer (DiT) for high-resolution video generation on a single Nvidia H200 GPU with 1.5 TB of host RAM, by ke…
Researchers propose DAMamba-UNet3D, a parameter-efficient state-space model U-Net for 3D medical image segmentation that achieves a mean Dice of 0.815 on BraTS 2020 with ~5.3 million parameters, roughly 13 times fewer th…
Researchers propose a three-stage pipeline that turns motion-imitation skills into a reusable hybrid motion prior (HMP) for humanoid locomotion, enabling task-level policies to be trained by selecting discrete codebook e…
A new study from arXiv proposes BettiSafe, a topology-guided split selection strategy for privacy-aware split learning that improves resistance to feature inversion attacks by 2 to 5 times compared to depth-based heurist…
Researchers introduced LC-SEPLM (Long-range Contact-supervised ESM Protein Language Model), which adapts ESM2 with LoRA and long-range residue-pair contact supervision while retaining sequence-only downstream inference. …
Researchers introduce Graded Large Language Models (GLLMs), an algebraic framework that adds a grading structure to transformer representation spaces, improving performance without increasing inference cost. The framewor…
A new AI framework called Collaborative Meta Knowledge Enhancement (COME) achieves state-of-the-art dementia etiology diagnosis with a mean macro-averaged AUC of 85.62%, outperforming the strongest baseline by 4.29 point…
A 2025 paper published at ICAIF 2025 and uploaded to arXiv on July 25, 2026, describes a compact neural architecture for portfolio optimization that reduces learnable parameters from 39,586 to 2,175. The authors report l…
Researchers propose Delta-InvFormer, a transformer-based surrogate model that uses visible-light video frames to predict two-dimensional light intensity distributions in tokamak divertors, accelerating traditional method…
A new framework for decentralized causal discovery using judo calculus, formally defined as j-stable causal inference with j-do-calculus in a topos of sheaves, achieves improved computational efficiency and performance o…
A new study introduces LENS, a protocol for evaluating whether machine-unlearning algorithms can suppress disinformation-aligned narrative frames in large language models. Testing on four near-12B multilingual models, th…
Researchers propose MIITA, a Memory-Induced Inference-Time Adaptation framework that enables small language models to perform continual learning under constrained storage without catastrophic forgetting. MIITA stores sup…
Researchers introduced WorldDiT, a unified diffusion transformer architecture that couples action generation with visual world modeling and achieves strong robot control performance without a large pretrained VLM action …
Researchers propose UNIFUSION, a continual pre-training approach that adapts pretrained autoregressive language models (e.g., GPT-2) to uniform-noise discrete diffusion under a unified reverse-rate objective. At 256 samp…
Researchers propose MXAttention, a data-free post-training quantization framework for MXFP4 attention that introduces Universal Optimal Scaling (UOS) with a distribution-independent optimal scaling boundary Qmax=7.25 and…
A team of researchers achieved a Codabench evaluation score of 8.40 in the TidyVoice 2026 Challenge by applying Nuisance Attribute Projection (NAP) as a simple language-normalization step for cross-lingual speaker verifi…