1

Learn large multimodal models with RLHF augmented by factual information to reduce hallucination.

Propose a multi-modal learning framework for 3D shapes with two learning stages and three unified adapter modules.

Learn to segment visual entities and their parts in an open-world with pure self-supervision.

Develop a fully self-supervised approach to object detection which understands part-to-whole composition and needs much less training costs.

Distill knowledge from an automatically designed sequence of teachers into a lightweight student object detector to mitigate the teacher-student capacity gap.

Unify contrastive learning for representation learning and Mean Teacher for domain adaptation into one general-purpose framework for learning unsupervised domain adaptive object detectors.

Analyze the slow convergence of DEtection TRansformer (DETR), and propose alternative solutions to improve convergence and performance of DETR.

Propose a novel neural architecture search formulation to encourage flatness in the architecture space, which improves generalization of searched architectures.

Search compressed network architectures in a learnable embedding space via Bayesian optimization, and find compressed networks outperforming hand-crafted architectures.