Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
A developer competing in the ARC Prize 2026 benchmark, which tests AI's ability to learn novel games from scratch, describes building a custom agent that hand-crafts perception and decision-making while using a language …
A developer has open-sourced UGR, an AI runtime operating system designed to orchestrate inference across commodity hardware by treating models as collections of resources that can move between memory tiers. The project …
Active learning is a machine learning approach where the model identifies which unlabeled data points are most uncertain and should be labeled by humans, reducing labeling costs and improving efficiency. Instead of rando…
Svetovid, a new WebGPU-powered tool, enables real-time visualization and debugging of PyTorch models, NumPy arrays, and large datasets directly in the browser. The tool supports interactive exploration of neural network …
Vector databases are a new type of database that stores data as vectors, enabling efficient similarity search for AI applications. Embeddings, which are numerical representations of objects, are used to capture semantic …
A free, ATS-friendly Data Scientist cover letter sample is available, featuring an AI generator that customizes the letter in one click. The sample highlights a candidate with expertise in predictive modeling, machine le…
A free, ATS-tested Data Scientist resume sample is available, featuring a customizable template with AI builder integration. The sample highlights a candidate with 4+ years of experience, including a customer churn predi…
Researchers propose UnMixNet, a physics-closed graph neural solver that embeds Maxwell-Stefan multicomponent transport PDEs, competitive adsorption ODEs, and nonlinear sensor transduction ODEs into end-to-end gas unmixin…
Researchers propose DuSPiT, a dual-branch sub-patch pixel diffusion transformer that separates global structural reasoning from local appearance modeling in pixel-space image generation, achieving richer details and a be…
A new study introduces LatentMT, the first systematic application of latent-reasoning looped language models (LoopLMs) to machine translation, achieving performance comparable to models three to five times larger using a…
Deep learning models for breast MRI tumor classification suffer near-chance external accuracy (0.5048–0.5265) when dataset origin is perfectly correlated with label, but a class-wise dataset mixing method using Duke Brea…
A new study from arXiv (2607.18695v1) finds that descriptors generated by large language models (LLMs) for zero-shot classification carry little visual evidence, collapsing ImageNet accuracy from 59.5% to 15.5% when the …
A new study from researchers at the University of Zurich finds that reinforcement learning with verifiable rewards surpasses supervised fine-tuning for legal neural machine translation, though enhanced small models still…
Researchers released four natural-language-to-MLIR benchmarks totaling 410 in-scope pairs and a three-layer schema-derived constraint stack that lets SmolLM2-1.7B match or exceed 15B-34B code LMs on structurally dominate…
A new method called domain-conditional position offsets reduces the cold-start penalty in autoregressive language models by adding a single learned vector to the embedding activation at the first sequence positions while…
Researchers at urbanAIthi have developed a preference-conditioned multi-objective reinforcement learning controller for transit signal priority (TSP) that allows runtime tuning of the trade-off between bus delay and over…
Researchers introduce Fluid-SDF, a differentiable Constructive Solid Geometry framework that models shapes using explicit geometric primitives with under 100 parameters, achieving comparable or superior intersection-over…
Researchers introduced E-SpecFormer (Edge Spectrum monitoring Transformer), an edge-efficient model for end-to-end automatic modulation and covert channel recognition. The Nano variant achieves 86.5% average accuracy on …
A deep learning ensemble developed from 128,621 CT examinations at nine Japanese institutions estimated adult sex, age, height, and weight from coronal digitally reconstructed radiographs with high accuracy, achieving 0.…
ECoNGS, an efficient compressive neural Gaussian splatting framework for volume visualization, improves reconstruction quality by up to 2.2 dB in PSNR while reducing model size by up to 6.1x and training time by up to 5.…