Picture for Tong Zhang

Tong Zhang

Nanjing University of Science and Technology, Nanjing, China

Rotated Runtime Smooth: Training-Free Activation Smoother for accurate INT4 inference

Add code
Sep 30, 2024
Figure 1 for Rotated Runtime Smooth: Training-Free Activation Smoother for accurate INT4 inference
Figure 2 for Rotated Runtime Smooth: Training-Free Activation Smoother for accurate INT4 inference
Figure 3 for Rotated Runtime Smooth: Training-Free Activation Smoother for accurate INT4 inference
Figure 4 for Rotated Runtime Smooth: Training-Free Activation Smoother for accurate INT4 inference
Viaarxiv icon

From Lists to Emojis: How Format Bias Affects Model Alignment

Add code
Sep 18, 2024
Figure 1 for From Lists to Emojis: How Format Bias Affects Model Alignment
Figure 2 for From Lists to Emojis: How Format Bias Affects Model Alignment
Figure 3 for From Lists to Emojis: How Format Bias Affects Model Alignment
Figure 4 for From Lists to Emojis: How Format Bias Affects Model Alignment
Viaarxiv icon

Data Augmentation via Latent Diffusion for Saliency Prediction

Add code
Sep 11, 2024
Figure 1 for Data Augmentation via Latent Diffusion for Saliency Prediction
Figure 2 for Data Augmentation via Latent Diffusion for Saliency Prediction
Figure 3 for Data Augmentation via Latent Diffusion for Saliency Prediction
Figure 4 for Data Augmentation via Latent Diffusion for Saliency Prediction
Viaarxiv icon

On the Limited Generalization Capability of the Implicit Reward Model Induced by Direct Preference Optimization

Add code
Sep 05, 2024
Figure 1 for On the Limited Generalization Capability of the Implicit Reward Model Induced by Direct Preference Optimization
Figure 2 for On the Limited Generalization Capability of the Implicit Reward Model Induced by Direct Preference Optimization
Figure 3 for On the Limited Generalization Capability of the Implicit Reward Model Induced by Direct Preference Optimization
Figure 4 for On the Limited Generalization Capability of the Implicit Reward Model Induced by Direct Preference Optimization
Viaarxiv icon

Building Math Agents with Multi-Turn Iterative Preference Learning

Add code
Sep 04, 2024
Figure 1 for Building Math Agents with Multi-Turn Iterative Preference Learning
Figure 2 for Building Math Agents with Multi-Turn Iterative Preference Learning
Figure 3 for Building Math Agents with Multi-Turn Iterative Preference Learning
Figure 4 for Building Math Agents with Multi-Turn Iterative Preference Learning
Viaarxiv icon

A Homogeneous Graph Neural Network for Precoding and Power Allocation in Scalable Wireless Networks

Add code
Aug 30, 2024
Figure 1 for A Homogeneous Graph Neural Network for Precoding and Power Allocation in Scalable Wireless Networks
Figure 2 for A Homogeneous Graph Neural Network for Precoding and Power Allocation in Scalable Wireless Networks
Figure 3 for A Homogeneous Graph Neural Network for Precoding and Power Allocation in Scalable Wireless Networks
Figure 4 for A Homogeneous Graph Neural Network for Precoding and Power Allocation in Scalable Wireless Networks
Viaarxiv icon

Learning-based Multi-View Stereo: A Survey

Add code
Aug 27, 2024
Figure 1 for Learning-based Multi-View Stereo: A Survey
Figure 2 for Learning-based Multi-View Stereo: A Survey
Figure 3 for Learning-based Multi-View Stereo: A Survey
Figure 4 for Learning-based Multi-View Stereo: A Survey
Viaarxiv icon

Localize-and-Stitch: Efficient Model Merging via Sparse Task Arithmetic

Add code
Aug 24, 2024
Viaarxiv icon

Practical Video Object Detection via Feature Selection and Aggregation

Add code
Jul 29, 2024
Viaarxiv icon

Traversing Pareto Optimal Policies: Provably Efficient Multi-Objective Reinforcement Learning

Add code
Jul 24, 2024
Viaarxiv icon