Gen Li

26

Papers

83

Total Citations

Papers (26)

Phantom: Subject-Consistent Video Generation via Cross-Modal Alignment

EgoGen: An Egocentric Synthetic Data Generator

EgoM2P: Egocentric Multimodal Multitask Pretraining

Advancing Dynamic Sparse Training by Exploring Optimization Opportunities

Accelerating Convergence of Score-Based Diffusion Models, Provably

Outlier Weighed Layerwise Sparsity (OWL): A Missing Secret Sauce for Pruning LLMs to High Sparsity

Adaptive Prototype Learning and Allocation for Few-Shot Segmentation

Towards High-Quality and Efficient Video Super-Resolution via Spatial-Temporal Data Overfitting

LOCATE: Localize and Transfer Object Parts for Weakly Supervised Affordance Grounding

OSRT: Omnidirectional Image Super-Resolution With Distortion-Aware Transformer

Image Super-Resolution Using Dense Skip Connections

SVD: A Large-Scale Short Video Dataset for Near-Duplicate Video Retrieval

VQFR: Blind Face Restoration with Vector-Quantized Dictionary and Parallel Decoder

VolumetricSMPL: A Neural Volumetric Body Model for Efficient Interactions, Contacts, and Collisions

Removing Interference and Recovering Content Imaginatively for Visible Watermark Removal

One-Shot Open Affordance Learning with Foundation Models

Nonconvex Low-Rank Tensor Completion from Noisy Data

Sample Complexity of Asynchronous Q-Learning: Sharper Analysis and Variance Reduction

Breaking the Sample Size Barrier in Model-Based Reinforcement Learning with a Generative Model

Sample-Efficient Reinforcement Learning Is Feasible for Linearly Realizable MDPs with Limited Revisiting

Breaking the Sample Complexity Barrier to Regret-Optimal Model-Free Reinforcement Learning

Minimax-Optimal Multi-Agent RL in Markov Games With a Generative Model

Reward-agnostic Fine-tuning: Provable Statistical Benefits of Hybrid Reinforcement Learning

Dynamic Sparsity Is Channel-Level Sparsity Learner

The Curious Price of Distributional Robustness in Reinforcement Learning with a Generative Model

Regret-Optimal Model-Free Reinforcement Learning for Discounted MDPs with Short Burn-In Time