Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
Baidu released Unlimited OCR, a 3B-parameter Mixture-of-Experts model that uses Reference Sliding Window Attention to keep KV cache constant, enabling efficient parsing of dozens of pages in one forward pass. The model s…
OpenAI's Whisper speech recognition model can run locally on a laptop CPU in real time, enabling offline dictation without sending audio to cloud servers. Off Grid AI Desktop is a free, open-source app that bundles whisp…
AI and modern tooling are reshaping geospatial work by enabling free satellite data, cloud platforms, and machine learning models to process imagery at scale. Practical workflows now use open-source libraries like geopan…
Egyptian fractions, a method of expressing rational numbers as sums of distinct unit fractions, were used in Egypt as early as 4000 BC and remained the primary way to write non-integer numbers in Europe until the 18th ce…
Researchers propose a novel neural network quantization method that learns quantization-aware linear paths in weight space to find low-loss subspaces, enabling direct quantization of the midpoint without performance degr…
Researchers propose G-SPIN, a graph-based framework that corrects phonetic errors in ASR outputs by combining a graph neural network for candidate generation with a masked language model and LLM for contextual re-ranking…
Researchers introduced OrthoTrack, a training-free system that estimates continuous 6-DoF UAV trajectories using public orthophotos and surface models, eliminating drift and GPS dependency. The system runs in real time o…
Researchers introduced MJEPA, a joint-embedding predictive architecture for audio-visual learning that uses a single unified encoder and a single predictive objective. The model outperforms prior frozen baselines by over…
Researchers propose Chorus II, a cross-request sparsity reuse framework for efficient image-to-video generation, achieving a 2.16× speedup by reusing sparse attention masks from similar historical requests. The method pr…
Researchers propose a hierarchical Bayesian framework for learning dynamical systems from multiple sparse datasets, modeling dataset-specific parameters as draws from a shared population distribution. The method embeds a…
Researchers introduced the Swarm-Inspired Emergent Synchronizer (SIES), a graph-dynamical framework that learns local interaction rules for collective behaviors across graphs, dynamics, and tasks. SIES achieved synchroni…
Researchers developed a reinforcement learning-based method for digital twin-driven bearing health monitoring that adaptively aligns simulated and real vibration data, achieving 92.8% cross-equipment diagnostic accuracy …
Researchers propose DCQ-GNN, a spectral graph neural network using convex-concave quadratic filters to improve spectral selectivity without high-order polynomials. The model achieves top average rank on heterophilic grap…
Researchers at arXiv propose a method to certify conservation laws in learned latent world models by decoding physical invariants from latent states. They derive shell-horizon certificates that decompose into representat…
Researchers propose a new approach to near-sensor computing that performs lightweight Neural Architecture Search (NAS) directly on deployment devices to adapt tiny neural networks to real-time sensor data. Validated on t…
Researchers used the AI-mathematician system AIM to co-discover sign-embedding quantum algorithms for matrix equations and matrix functions, advancing quantum linear algebra. The human-AI collaboration transformed a vagu…
Researchers introduced block-sparse featurizers (BSFs) to capture visual concept manifolds in neural network representations, showing that concepts are low-dimensional geometric structures rather than isolated directions…
A new survey from arXiv redefines continual learning for large language models as a closed-loop update-and-release problem within a versioned ecosystem, identifying three core challenges and five lifecycle design princip…
Researchers propose SEMIR, a framework that replaces the pixel lattice with a parameterized graph minor to preserve thin-structure connectivity in segmentation tasks. The method achieves full-resolution inference without…
Researchers introduced Local Branch Routing (LBR), a token-level test-time scaling framework that expands a small local lookahead tree and uses a lightweight router to select the best branch, enabling efficient and train…