Statistical Learning
BoostAdapter: Improving Vision-Language Test-Time Adaptation via Regional Bootstrapping
Adaptation of pretrained vision-language models such as CLIP to various downstream tasks have raised great interest in recent researches.
BLoB: Bayesian Low-Rank Adaptation by Backpropagation for Large Language Models Yibin Wang
Current research predominantly focuses on eliciting the internal capability of uncertainty estimation of LLMs.
Subwords as Skills: Tokenization for Sparse-Reward Reinforcement Learning David Y unis TTI-Chicago Chicago, IL
Exploration in sparse-reward reinforcement learning (RL) is difficult due to the need for long, coordinated sequences of actions in order to achieve any reward.
Zero-Shot Transfer of Neural ODEs
This challenge necessitates new mathematical and algorithmic tools that enable adaptation and zero-shot transfer.
Truthful High Dimensional Sparse Linear Regression
The problem involves two intertwined components: data acquisition and privacy-preserving data analysis.
