每天自动聚合 AI 领域最新动态
When asked about entities outside their knowledge boundary, LLMs routinely fabricate plausible-sounding details rathe...
This report presents an improved version of AlayaWorld. While the backbone architecture, chunk-wise autoregressive ge...
Current large language model development relies on massive, often non-permissible datasets, creating a high barrier f...
We study masking diffusion for discrete sampling and introduce a path-resolved measure of data geometry called the \e...
AI agents are increasingly used for programming, but do not provide any guarantee on the correctness of generated cod...
Modern language models are trained on heterogeneous web-scale text corpora. Consequently, studying knowledge and skil...
LLM coding agents issue Bash commands through interfaces that may serialize, wrap, and reparse model output. Matched ...
Humanoid motion tracking is central to teleoperation and whole-body imitation, yet evaluation often disagrees with wh...
Recent advances in foundation models have enabled AI scientists to automate increasingly complete research workflows,...
Transforming multimodal sources into condensed and structured media outputs can be fundamentally conceptualized as a ...
Autoregressive transformers compute along two axes: horizontally across generated tokens, and vertically through mode...
Video world models simulate future states conditioned on current observations and user actions. Recent systems have d...