3 papers
cs.LG2025
Re-Evaluating the Impact of Unseen-Class Unlabeled Data on Semi-Supervised Learning Model
Rundong He, Yicong Dong, Lanzhe Guo +2
Semi-supervised learning (SSL) effectively leverages unlabeled data and has been proven successful across various fields. Current safe SSL methods believe that unseen classes in un…
cs.AI2024
A Moonshot for AI Oracles in the Sciences
Bryan Kaiser, Tailin Wu, Maike Sonnewald +2
Nobel laureate Philip Anderson and Elihu Abrahams once stated that, "even if machines did contribute to normal science, we see no mechanism by which they could create a Kuhnian rev…
cs.LG2023
How Well Does GPT-4V(ision) Adapt to Distribution Shifts? A Preliminary Investigation
Zhongyi Han, Guanglin Zhou, Rundong He +7
In machine learning, generalization against distribution shifts -- where deployment conditions diverge from the training scenarios -- is crucial, particularly in fields like climat…