Picture for Chao Ma

Chao Ma

HaWoR: World-Space Hand Motion Reconstruction from Egocentric Videos

Add code
Jan 06, 2025
Viaarxiv icon

SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training

Add code
Dec 23, 2024
Figure 1 for SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training
Figure 2 for SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training
Figure 3 for SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training
Figure 4 for SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training
Viaarxiv icon

Breaking the Context Bottleneck on Long Time Series Forecasting

Add code
Dec 21, 2024
Viaarxiv icon

Evaluating and Enhancing LLMs for Multi-turn Text-to-SQL with Multiple Question Types

Add code
Dec 21, 2024
Viaarxiv icon

Cross-View Consistency Regularisation for Knowledge Distillation

Add code
Dec 21, 2024
Viaarxiv icon

SWAN: Preprocessing SGD Enables Adam-Level Performance On LLM Training With Significant Memory Reduction

Add code
Dec 17, 2024
Figure 1 for SWAN: Preprocessing SGD Enables Adam-Level Performance On LLM Training With Significant Memory Reduction
Figure 2 for SWAN: Preprocessing SGD Enables Adam-Level Performance On LLM Training With Significant Memory Reduction
Figure 3 for SWAN: Preprocessing SGD Enables Adam-Level Performance On LLM Training With Significant Memory Reduction
Figure 4 for SWAN: Preprocessing SGD Enables Adam-Level Performance On LLM Training With Significant Memory Reduction
Viaarxiv icon

OccScene: Semantic Occupancy-based Cross-task Mutual Learning for 3D Scene Generation

Add code
Dec 15, 2024
Viaarxiv icon

Heterogeneous Graph Transformer for Multiple Tiny Object Tracking in RGB-T Videos

Add code
Dec 14, 2024
Viaarxiv icon

PhyCAGE: Physically Plausible Compositional 3D Asset Generation from a Single Image

Add code
Nov 27, 2024
Viaarxiv icon

Teaching Video Diffusion Model with Latent Physical Phenomenon Knowledge

Add code
Nov 18, 2024
Viaarxiv icon