Picture for Han Liu

Han Liu

Johns Hopkins University

On Statistical Rates of Conditional Diffusion Transformers: Approximation, Estimation and Minimax Optimality

Add code
Nov 26, 2024
Figure 1 for On Statistical Rates of Conditional Diffusion Transformers: Approximation, Estimation and Minimax Optimality
Figure 2 for On Statistical Rates of Conditional Diffusion Transformers: Approximation, Estimation and Minimax Optimality
Figure 3 for On Statistical Rates of Conditional Diffusion Transformers: Approximation, Estimation and Minimax Optimality
Figure 4 for On Statistical Rates of Conditional Diffusion Transformers: Approximation, Estimation and Minimax Optimality
Viaarxiv icon

Transformers are Deep Optimizers: Provable In-Context Learning for Deep Model Training

Add code
Nov 25, 2024
Viaarxiv icon

Fundamental Limits of Prompt Tuning Transformers: Universality, Capacity and Efficiency

Add code
Nov 25, 2024
Viaarxiv icon

One-Layer Transformer Provably Learns One-Nearest Neighbor In Context

Add code
Nov 16, 2024
Viaarxiv icon

On Differentially Private String Distances

Add code
Nov 08, 2024
Viaarxiv icon

Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent

Add code
Nov 05, 2024
Figure 1 for Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent
Figure 2 for Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent
Figure 3 for Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent
Figure 4 for Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent
Viaarxiv icon

Statistical Guarantees for Lifelong Reinforcement Learning using PAC-Bayesian Theory

Add code
Nov 01, 2024
Viaarxiv icon

Global Convergence in Training Large-Scale Transformers

Add code
Oct 31, 2024
Figure 1 for Global Convergence in Training Large-Scale Transformers
Figure 2 for Global Convergence in Training Large-Scale Transformers
Viaarxiv icon

Provably Optimal Memory Capacity for Modern Hopfield Models: Transformer-Compatible Dense Associative Memories as Spherical Codes

Add code
Oct 30, 2024
Figure 1 for Provably Optimal Memory Capacity for Modern Hopfield Models: Transformer-Compatible Dense Associative Memories as Spherical Codes
Figure 2 for Provably Optimal Memory Capacity for Modern Hopfield Models: Transformer-Compatible Dense Associative Memories as Spherical Codes
Figure 3 for Provably Optimal Memory Capacity for Modern Hopfield Models: Transformer-Compatible Dense Associative Memories as Spherical Codes
Figure 4 for Provably Optimal Memory Capacity for Modern Hopfield Models: Transformer-Compatible Dense Associative Memories as Spherical Codes
Viaarxiv icon

Sequential LLM Framework for Fashion Recommendation

Add code
Oct 15, 2024
Figure 1 for Sequential LLM Framework for Fashion Recommendation
Figure 2 for Sequential LLM Framework for Fashion Recommendation
Figure 3 for Sequential LLM Framework for Fashion Recommendation
Figure 4 for Sequential LLM Framework for Fashion Recommendation
Viaarxiv icon