每天自动聚合 AI 领域最新动态
Recovering an editable design file from a raster image is a common and costly bottleneck in modern design workflows, ...
We present a reproducible pipeline for mapping Common Vulnerabilities and Exposures (CVEs) to MITRE ATT&CK Enterprise...
Multimodal automated fact-checking (MAFC) verifies claims by retrieving and reasoning over external evidence. However...
On-policy distillation (OPD) grounds token-level supervision in the student's own trajectory, yet suffers from prefix...
Any-to-any models predict any modality from any combination of others within a single network, a formulation used in ...
Generation in video diffusion or flow models is computationally expensive due to the slow and iterative sampling proc...
Joint-Embedding Predictive Architectures (JEPAs) learn world models by predicting in representation space rather than...
Different research lines use the term world model in different ways, yet they share a common aim: to capture how the ...
Compressed short-text generators can fail in two different places: the codec may discard information before generatio...
Parametric retrieval enables LLMs to retrieve tools implicitly by assigning each API a unique virtual token and train...
We introduce a vocabulary for automated research systems built from one or more agents to make their design choices e...
Many recent robot policies pursue stronger control by using large pretrained vision-language models (VLMs) as the act...