每天自动聚合 AI 领域最新动态
The rapid progression of large language models is extending AI from passive content generation into the active workfl...
Block-based video codecs select coding parameters based on the input by optimizing a rate-distortion trade-off. The c...
Recent advances in large language models (LLMs) have enabled agents to tackle long-horizon tasks across diverse envir...
Object permanence and solidity are hallmarks of human cognitive priors. Recent studies show that video generation mod...
Scientific discovery begins where known problems end. There, AI systems must engage in exploration: framing hypothese...
Video generation begins in text space by authoring a cinematic screenplay, then materializes into pixels. As contempo...
General-purpose vision-language models (VLMs) bring broad knowledge and spatial reasoning to robot manipulation, yet ...
Humans can effortlessly localize the direction of a sound source and integrate it with visual cues for reasoning, yet...
We introduce PUBG Ally, an embodied agent for PUBG: BATTLEGROUNDS that can reason, act autonomously, and play alongsi...
Modern AI agents routinely cross trust boundaries: they ingest untrusted content, combine it with privileged instruct...
Few-step autoregressive (AR) video diffusion enables low-latency streaming generation, but existing post-training met...
Deep search requires LLM agents to decompose complex queries, search for evidence, and synthesize grounded answers, y...