Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
Researchers propose a framework to expand human motion generation by using large-scale synthetic motion data and a redesigned VQ-VAE tokenizer, overcoming the limited diversity of motion capture datasets. The approach im…
Researchers propose automation-in-design (AiD) using deep learning and generative AI to automate high-tech system design, demonstrated through e-drive and spatial dimensioning case studies. The framework aims to shift en…
Researchers propose a new RANSAC scoring method that eliminates the need for a user-supplied inlier scale parameter by marginalizing it analytically under a conjugate Inverse-Gamma prior. The method achieves state-of-the…
Researchers introduced SelectAnyTree, a promptable instance segmentation model for 3D forest LiDAR point clouds that delineates individual trees from a few clicks. The model achieves 78.2% Intersection over Union (IoU) f…
Researchers propose SD-GPS, a solver-driven framework for geometry problem solving that integrates autoformalization and theorem proposing. The framework uses a symbolic solver as an execution oracle to improve formaliza…
Researchers propose HIA-GAT, a heterogeneous graph attention network for frame-level traffic conflict risk prediction on freeways. The model outperforms baselines on NGSIM datasets, achieving AUC up to 0.867, and provide…
Researchers developed a geometry-conditioned Fourier neural operator (FNO) to approximate the solution operator for the cubic nonlinear Schrödinger equation on two-dimensional tori with varying aspect ratios. The model c…
Researchers propose EntMTP, a training-free scheduler that dynamically adjusts multi-token prediction depth based on local entropy, achieving up to 1.36x speedup over Medusa baselines in LLM inference without quality los…
Researchers introduced Tessellating the Earth (TTE), a learnable location encoder using Spherical Voronoi partitions that allocates representational capacity to discriminative areas. TTE sets a new state of the art in ge…
Researchers propose Transformation-Aware Decoupling (TAD), a framework for 3D scene graph generation that improves robustness to viewpoint changes by decoupling relation reasoning into stable and directional components. …
Researchers propose ATOD, a hybrid online distillation algorithm that combines on-policy distillation and reinforcement learning to train small language-model agents for multi-turn tasks. ATOD uses an annealed schedule t…
Researchers introduced KARMA, a method for explaining multivariate time series forecasting models by constructing a Markov surrogate model that captures temporal dependencies. The approach identifies minimal predictive h…
Researchers propose HybridCodec, a novel approach combining discrete tokens with continuous residuals to improve speech representation in language models, reducing information loss and autoregressive steps while retainin…
Carles Marin released an open-source bilingual guide on Transformer architectures, covering attention dynamics, KV-cache compression, and advanced concepts like grokking. The resource includes reproducible code and inter…
Researchers have developed a method for on-policy distillation (OPD) that works across different tokenizers, enabling knowledge transfer between large language models (LLMs) from different model families. The approach us…
A developer released an open-source repository for polygenic prediction and embryo selection, enabling users to compute polygenic scores for traits like intelligence and height using public genome-wide association study …
Tau is an educational Python coding agent designed to teach how agents are built by exposing every component—provider adapters, harness, coding environment, and terminal UI—in readable layers. The project emphasizes a cl…
An independent research project analyzed the internal dynamics of small and medium-sized language models, revealing that functional properties become linearly decodable in hidden representations and that models cluster i…
A developer outlines practical patterns for building multi-agent systems with Python, including the ReAct architecture, tool selection, memory hierarchy, and error recovery. The guide emphasizes starting with narrow task…
A Seoul-based backend developer built k-cosmos, an interactive 3D music space that maps K-pop tracks using 768-dimensional vector embeddings, running entirely on the OCI Free Tier. The system uses a self-reinforcing pipe…