5 papers
CoGe-GCD: Reframing Generalized Category Discovery with Compositional Generalization
Luyao Tang, Jiewei Zheng, Kunze Huang +3
Generalized Category Discovery (GCD) assigns unlabeled instances, mixed with labeled data, to known or novel categories, requiring human-like compositional reasoning: reusing primi…
OCRT: Boosting Foundation Models in the Open World with Object-Concept-Relation Triad
Luyao Tang, Yuxuan Yuan, Chaoqi Chen +3
Although foundation models (FMs) claim to be powerful, their generalization ability significantly decreases when faced with distribution shifts, weak supervision, or malicious atta…
Bootstrap Segmentation Foundation Model under Distribution Shift via Object-Centric Learning
Luyao Tang, Yuxuan Yuan, Chaoqi Chen +3
Foundation models have made incredible strides in achieving zero-shot or few-shot generalization, leveraging prompt engineering to mimic the problem-solving approach of human intel…
Mixstyle-Entropy: Domain Generalization with Causal Intervention and Perturbation
Luyao Tang, Yuxuan Yuan, Chaoqi Chen +2
Despite the considerable advancements achieved by deep neural networks, their performance tends to degenerate when the test environment diverges from the training ones. Domain gene…
Activate and Reject: Towards Safe Domain Generalization under Category Shift
Chaoqi Chen, Luyao Tang, Leitian Tao +4
Albeit the notable performance on in-domain test points, it is non-trivial for deep neural networks to attain satisfactory accuracy when deploying in the open world, where novel do…