Multi-object tracking (MOT) is a fundamental problem in computer vision ...
The pre-training-fine-tuning paradigm based on layout-aware multimodal
p...
To bring digital avatars into people's lives, it is highly demanded to
e...
We present a reinforcement learning (RL) framework that enables quadrupe...
Human motion synthesis is a long-standing problem with various applicati...
Dialogue policy learning, a subtask that determines the content of syste...
This paper presents our task-oriented dialog system UBAR which models
ta...