Mamba is a novel architecture thast addresses Transformers’ computational inefficiency on long sequences.

A proposed new causal inference dataset with a bespoke chain-or-thought prompting strategy.

EnCLAP is a novel framework for automated caption, utilizing two acoustic representation models with a pretrained language model.

PostGPT is a multi-modal LLM for pose generation and esimation.

An algorithm that captures the soft constraints with some expert demonstrations.

An algorithm for simultaneously estimating multiple constraints corresponding to different types of experts.

Introducing an ICRL benchmark in the context of RL application domains.

An uncertainty-aware Reinforcement Learning (RL) framework to learn a risk-sensitive player evaluation metric from stochastic game dynamics.

A method that alternates single-agent IRL to learn a reward function for multiple agents.

A survey of deep Keyword Spotting Systems. 10

A survey overview of the task audio captioning. 11

An overview of the future directions in Anomalous SED. 12