News: Our Group is Heading to CVPR!

We are thrilled to announce that our research group will have a massive presence at the upcoming Conference on Computer Vision and Pattern Recognition (CVPR)!

Our team will be presenting a vast range of cutting-edge research, spanning 3D scene encoding, vision-language models, adversarial robustness, and robotic manipulation. If you are attending CVPR, we would love to connect with you.

Check out our full presentation schedule below, come by our posters, and say hi!

📅 Presentation Schedule

Friday, June 5

• 07:00 - 08:30 | Finetune Like You Pretrain: Boosting Zero-shot Adversarial Robustness in Vision-language Models
  • Session: Findings 1 | Poster #: 71
• 10:45 - 12:45 | From Weights to Concepts: Data-Free Interpretability of CLIP via Singular Vector Decomposition
  • Session: 1 | Poster #: 267
• 10:45 - 12:45 | The Devil Is in Gradient Entanglement: Energy-Aware Gradient Coordinator for Robust Generalized Category Discovery
  • Session: 1 | Poster #: 329
• 16:00 - 18:00 | CLIPoint3D: Language-Grounded Few-Shot Unsupervised 3D Point Cloud Domain Adaptation
  • Session: 2 | Poster #: 224
• 16:00 - 18:00 | Token Reduction via Local and Global Contexts Optimization for Efficient Video Large Language Models
  • Session: 2 | Poster #: 305

Saturday, June 6

• 11:45 - 13:45 | Open-Vocabulary Domain Generalization in Urban-Scene Segmentation
  • Session: 3 | Poster #: 574
• 11:45 - 13:45 | PoInit-of-View: Poisoning Initialization of Views Transfers Across Multiple 3D Reconstruction Systems
  • Session: 3 | Poster #: 600
• 11:45 - 13:45 | TerraScope: Pixel-Grounded Visual Reasoning for Earth Observation
  • Session: 3 | Poster #: 230
• 16:45 - 18:45 | Chorus: Multi-Teacher Pretraining for Holistic 3D Gaussian Scene Encoding
  • Session: 4 | Poster #: 1
• 16:45 - 18:45 | Generalizable Knowledge Distillation from Vision Foundation Models for Semantic Segmentation
  • Session: 4 | Poster #: 487

Sunday, June 7

• 07:30 - 09:00 | Large Multimodal Models as General In-Context Classifiers
  • Session: Findings 3 | Poster #: 11
• 07:30 - 09:00 | Organizing Unstructured Image Collections using Natural Language
  • Session: Findings 3 | Poster #: 227
• 07:30 - 09:00 | SEM: Sparse Embedding Modulation for Post-Hoc Debiasing of Vision-Language Models
  • Session: Findings 3 | Poster #: 147
• 11:45 - 13:45 | Deconstructing the Failure of Ideal Noise Correction: A Three-Pillar Diagnosis
  • Session: 5 | Poster #: 549
• 11:45 - 13:45 | How to Take a Memorable Picture? Empowering Users with Actionable Feedback
  • Session: 5 | Poster #: 104
• 15:30 - 17:30 | Hyperbolic Busemann Neural Networks
  • Session: 6 | Poster #: 544
• 15:30 - 17:30 | Specificity-aware reinforcement learning for fine-grained open-world classification
  • Session: 6 | Poster #: 491

See you in person or online! Follow our research group on social media for live updates directly from the conference floor.