每天自动聚合 AI 领域最新动态
World-Action Models inherit world knowledge from video-generative priors, and channel it into executable control sign...
We present ENEAS, a unified, text-promptable method for instance tracking and semantic discovery. Text-promptable seg...
Safety alignment is usually posed as a topic-level question: is this subject harmful? Deployments ask a narrower one....
Causal inference is the practice of estimating the effect of a treatment or intervention from data. It traditionally ...
Many NLP tasks, such as summarization and extractive question answering, reduce to retrieving relevant content from d...
We present a systems-security case study of a two-node split-LLM training system whose privacy evaluation passed whil...
Change data synthesis provides a cost-effective solution for expanding training data and improving the performance of...
On-policy distillation (OPD) accelerates post-training by providing dense token-level supervision from a frozen teach...
Large Language Models (LLMs) often help users generate artifacts through iterative cycles of generation and revision ...
A reasoning model can improve from its own on-policy experience, but this inner loop is fragile: terminal verifiers p...
Vision-language-action (VLA) models map visual observations and language instructions directly to robot actions, but ...
On-policy distillation trains a language model on its own generations while a teacher scores them token by token. It ...