Title of the paper
Two-line summary of the contribution and scope.
Use thematic view for a quick scan and chronological view for a full timeline. Full list on Google Scholar.
For hiring and collaboration discussions, the first two thematic sections are the best starting point.
Two-line summary of the contribution and scope.
A taxonomy for evaluating hierarchical scene reasoning in vision-language models.
Joint 3D part segmentation and semantic naming.
Part segmentation learned from synthetic animal data.
A comprehensive dataset for 3D animal pose and shape.
Shared weight-subspace structure for efficient adaptation and continual learning.
Resource-efficient adaptation and inference via adapter recycling.
A shared-subspace approach toward almost strict continual learning for large models.
Continual learning with optimal relevance mapping.
Incremental neural mesh models for class-incremental learning.
Adaptive neural connectivity for sparsity-aware learning.
Scaling compositional models for robust 3D classification and pose.
Source-free domain adaptation for category-level pose estimation.
Bayesian OOD robustness in image classification.
Fast 4D generation through diffusion-based triplane re-posing.
Pose-free sparse-view scene reconstruction using diffusion priors.
Improved alignment in text-to-image generative models.
Offline outdoor navigation with full privacy.
Timing attack analysis on AES on modern processors.
Real-time neural model-based human detection and behavior classification.
Joint 3D part segmentation and semantic naming.
A shared-subspace approach toward almost strict continual learning for large models.
A taxonomy for evaluating hierarchical scene reasoning in vision-language models.
Shared weight-subspace structure for efficient adaptation and continual learning.
Scaling compositional models for robust 3D classification and pose.
Fast 4D generation through diffusion-based triplane re-posing.
Pose-free sparse-view scene reconstruction using diffusion priors.
Improved alignment in text-to-image generative models.
Resource-efficient adaptation and inference via adapter recycling.
Incremental neural mesh models for class-incremental learning.
Part segmentation learned from synthetic animal data.
Source-free domain adaptation for category-level pose estimation.
Bayesian OOD robustness in image classification.
A comprehensive dataset for 3D animal pose and shape.
Continual learning with optimal relevance mapping.
Adaptive neural connectivity for sparsity-aware learning.
Real-time neural model-based human detection and behavior classification.
Offline outdoor navigation with full privacy.
Timing attack analysis on AES on modern processors.