News
AI, cybersecurity, machine learning, scraping & programming news.
105 results on this page · clear filters
Breakdown of Local Denoising as Semantic Speciation
The dynamics of generative models exhibit two apparently distinct temporal windows: a speciation window, in which a sample commits to a semantic class, and a nonlocality window, in which local context windows become insufficient f...
LeapQuant: Efficient Linear Attention with Accurate Recurrent State Quantization
Recent large language models increasingly adopt hybrid attention designs that replace the standard quadratic attention with linear alternatives. Models such as Gated DeltaNet (GDN) and Kimi Delta Attention (KDA) compress the entir...
Breaking the Uniformity Trap: Scaling Video Diffusion Model via SplitMoE
Video diffusion models have achieved remarkable fidelity in text-to-video synthesis, yet scaling these models to capture complex spatiotemporal dynamics remains computationally expensive. Diffusion Transformers (DiTs) have become...
Multi-Agent Flow Matching with Decoupled Generative Guidance
Generative modeling has become a cornerstone for producing diverse objects from complex, multimodal distributions. Diffusion models and flow matching frameworks have demonstrated remarkable success across domains ranging from lang...
PDMD: Projected Distribution Matching Distillation for Video Diffusion Models
Modern video diffusion models require many denoising evaluations over long spatiotemporal token sequences, making inference computationally expensive. Distribution Matching Distillation (DMD) was introduced to reduce the number of...
Unifying Distributional Training for One-Step Visual Generation
One-step visual generation requires efficient supervision that can shape a generator in a single forward pass. Prior distributional training methods match real and generated features in frozen representation spaces, but lack a uni...
Improving Test-Time Scaling with Adaptive Looped Transformers
Looped transformers have emerged as a parameter-efficient architecture by reusing a shared stack of layers across multiple input positions during a single forward pass. This reuse treats computation as a dynamic process that can b...
Copy the Same, Distill the Difference: Initializing Linear Vision Transformers
Linear Vision Transformers (ViTs) aim to replace the softmax attention mechanism with a linear-complexity alternative, enabling more efficient token routing at scale. However, a persistent obstacle is that linear ViTs typically re...
New LoRA Skills Should Read but Never Write
Large language models implicitly infer attributes of their users and adjust their behavior accordingly, yet these internal user models remain difficult to inspect and causally manipulate. A team from the Technical University of Mu...
Machine Learning Systems
The article is ~746 words. Here it is in HTML format: A new textbook from Harvard University is reframing how engineers think about machine learning systems. Titled "Machine Learning Systems" and subtitled "The physics of AI engin...