Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
Researchers developed the Language-guided Segmentation-assisted Diagnostic Transformer (LSDT) model, which uses zero-shot segmentation and multi-modal slice fusion to classify prostate cancer into four clinically meaning…
Researchers introduce FogDrive, a multi-modal synthetic driving dataset for autonomous driving under fog, containing 660 scenes (~133k frames) with four synchronized cameras, LiDAR, and radar across three calibrated fog …
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 approach to Hardware Trojan detection uses And-Inverter Graphs (AIGs) and Knowledge Graph Embedding (KGE) to scale to modern System-on-Chips with billions of gates, achieving linear complexity and clear geometric s…
Researchers introduced the Multimodal Interactive Motion Encoder (MIME), the first dedicated multimodal encoder for two-person interactive motion, achieving a 12.8% relative improvement in text-to-motion R@1 on the Inter…
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…
EchoBridge, a new model for aligning ECG and echocardiography data, improves classifier-free AUROC, AUPRC, and F1 by 7.88, 5.61, and 4.54 points over baselines, according to a preprint on arXiv. The model uses Complement…
Researchers introduced PatiGonit22K, a dataset of 22,441 Bengali mathematical word problems designed to advance natural language understanding and quantitative reasoning in low-resource languages. The dataset extends the…
Researchers have developed the first patient-independent multimodal depression detection framework incorporating domain generalization, achieving 93.2% accuracy and 94.2% F1-score on the Androids-Corpus dataset. The mode…
Researchers propose a reference feature atlas for auditing language models, trained once on a reference panel and reused for new targets by fitting only a linear decoder. In tests on five 7-9B instruction-tuned models an…
Researchers propose FMOPF, a framework using latent flow matching and constraint-aware interaction priors to generate diverse near-optimal solutions for AC optimal power flow, addressing the limitations of existing diffu…
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 introduce TOM-GS, an editable video representation that uses regular 3D Gaussians with a continuous temporal opacity formulation, avoiding complex deformations. The method outperforms prior editable video rep…
Researchers introduced CausalGate, an intervention-guided framework for transformer module pruning that measures semantic damage via Kullback-Leibler divergence, outperforming dynamic routing baselines on TinyLlama-1.1B,…
A study using EEG signals and a custom convolutional neural network (CNN) predicted the outcome of repetitive transcranial magnetic stimulation (rTMS) therapy for major depressive disorder (MDD) with 93.60% accuracy, acc…
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 …
A new study from arXiv identifies a fundamental limitation in standard attention mechanisms: when evidence patterns repeat, the numerator and denominator grow at the same rate, causing inputs with different amounts of ac…
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…
Researchers propose Co-E, a training-free system that synchronizes graph and text working memory for multi-hop question answering, outperforming comparable open-backbone baselines on six benchmarks and rivaling larger or…
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…