每天自动聚合 AI 领域最新动态
Long-horizon LLM agents must preserve information from past interactions to support future tasks. Existing memory sys...
Transforming multimodal sources into condensed and structured media outputs can be fundamentally conceptualized as a ...
We present the first systematic study of Massive activations (MAs) in layer-interleaved HLA LLMs and uncover two arch...
An LLM agent's capability depends not only on model weights but on its harness: prompts, tools, skills, and control f...
Spatial intelligence is becoming a foundation for embodied agents, robotic planning, and multimodal assistants. To im...
We present DreamX-Phi 1.0, an action-conditioned video world model for robotic manipulation that, given an observed f...
No single large language model (LLM) is optimal across all queries and budget constraints, making model routing essen...
Large-scale manipulation data is essential for robot learning, yet collecting robot demonstrations remains expensive ...
Semi-supervised semantic segmentation has long turned on one question, which pseudo-labels to trust, and a generation...
Pose-driven human animation synthesizes a video of a target person from a single reference image and a driving pose s...
Talking-video character replacement requires coordinated transfer of appearance and voice while preserving the source...
Agent skills represent a standardized format for packaging procedural knowledge and domain expertise, serving within ...