每天自动聚合 AI 领域最新动态
Game generation is an emerging application of coding agents, requiring models to transform natural-language specifica...
Reinforcement learning with verifiable rewards (RLVR) improves language-model reasoning, but GRPO-style optimization ...
Deep research agents are increasingly evaluated on their ability to search for evidence, reason over retrieved source...
Memory has become a standard substrate for self-evolving agents, yet retaining experience is not the same as learning...
Knowledge distillation transfers a teacher's competence to a small student but is brittle in the small-student regime...
Looped Transformers scale latent computation by repeatedly applying shared blocks, but sequential looping increases l...
Training computer-use agents (CUAs) -- models that interact with graphical desktops through screenshots and keyboard/...
Scaling model size, specifically depth and width, has driven significant progress in transformer-based language model...
Vision-Language-Action (VLA) models benefit from large-scale and diverse embodied data, yet scaling robot trajectory ...
Pixel-space diffusion models are trained on full-bandwidth noisy images, yet the useful signal available to the denoi...
Generating realistic humanoid motion from scene images and text involves both low-frequency pose semantics and high-f...
Large language models perform increasingly well on standardized logical reasoning benchmarks, but whether this abilit...