Dong Gong

Assistant Professor@University of New South Wales·澳大利亚·OpenReview

研究方向

diffusion model · language model · OOD detection · continual learning · Semi-supervised learning · unsupervised/semi-supervised learning · memory networks · anomaly detection · deep learning · Image Restoration · Computer Vision · image restoration · low-level vision · computer vision

Dong Gong

Beyond DAGs: A Latent Partial Causal Model for Multimodal Learning

I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data?

Concept Component Analysis: A Principled Approach for Concept Extraction in LLMs

Adaptive Rank, Reduced Forgetting: Knowledge Retention in Continual Learning Vision-Language Models with Dynamic Rank-Selective LoRA

Little By Little: Continual Learning via Self-Activated Sparse Mixture-of-Rank Adaptive Learning

Towards Identifiable Latent Additive Noise Models

PiLoRA: Gradient-Informed Parameter-Importance-Aware Low-Rank Adaptation

Analytic DAG Constraints for Differentiable DAG Learning

FlashMo: Geometric Interpolants and Frequency-Aware Sparsity for Scalable Efficient Motion Generation

Model Inversion with Layer-Specific Modeling and Alignment for Data-Free Continual Learning

Coreset Selection via Reducible Loss in Continual Learning

Mining your own secrets: Diffusion Classifier Scores for Continual Personalization of Text-to-Image Diffusion Models

Beyond DAGs: A Latent Partial Causal Model for Multimodal Learning

Turning Challenges into Opportunities: How Distribution Shifts Enhance Identifiability in Causal Representation Learning

Learning Mamba as a Continual Learner

Tokens on Demand: Token Condensation as Training-free Test-time Adaptation