每天自动聚合 AI 领域最新动态
Robotic bin packing requires long-horizon sequential decision-making, as each object placement affects the available ...
Video is a rich representation of a physical event, capturing appearance, geometry, motion, and temporal evolution. O...
Representation Autoencoders (RAEs) enable diffusion models to operate in the feature spaces of pretrained visual enco...
Learned-memory methods store information in an explicit table and consume it through a separate reader, allowing addr...
Scalable simulation is essential for robot data generation, policy training, evaluation, and safe iteration, yet real...
Pruning large pre-trained transformer-based ASR models such as OpenAI's Whisper has seen great adoption, as pruning t...
Transformer architectures exhibit cross-layer redundancies, yet post-training compression pipelines typically optimiz...
Solutions based on large language models (LLMs) often rely on temperature sampling to improve accuracy and stability ...
Calibration of language models -- the alignment between expressed or implicit confidence and empirical correctness --...
Can one language model hand its live memory to another without the receiver rereading the context? We demonstrate use...
LLM-as-a-judge enables evaluation across diverse tasks, but inference cost and confidence reliability become critical...
A multi-agent system can reduce latency on complex tasks by executing work concurrently. Several pioneering harness f...