Picture for Zhangyang Wang

Zhangyang Wang

Atlas

Finding Fantastic Experts in MoEs: A Unified Study for Expert Dropping Strategies and Observations

Add code
Apr 10, 2025
Viaarxiv icon

More is Less: The Pitfalls of Multi-Model Synthetic Preference Data in DPO Safety Alignment

Add code
Apr 03, 2025
Viaarxiv icon

Can Test-Time Scaling Improve World Foundation Model?

Add code
Mar 31, 2025
Viaarxiv icon

Feature4X: Bridging Any Monocular Video to 4D Agentic AI with Versatile Gaussian Feature Fields

Add code
Mar 26, 2025
Viaarxiv icon

An Overview of Low-Rank Structures in the Training and Adaptation of Large Models

Add code
Mar 25, 2025
Viaarxiv icon

SPIN-Bench: How Well Do LLMs Plan Strategically and Reason Socially?

Add code
Mar 16, 2025
Viaarxiv icon

Make Optimization Once and for All with Fine-grained Guidance

Add code
Mar 14, 2025
Viaarxiv icon

You Only Debias Once: Towards Flexible Accuracy-Fairness Trade-offs at Inference Time

Add code
Mar 10, 2025
Viaarxiv icon

Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam

Add code
Feb 24, 2025
Figure 1 for Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam
Figure 2 for Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam
Figure 3 for Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam
Figure 4 for Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam
Viaarxiv icon

MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models

Add code
Feb 20, 2025
Figure 1 for MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models
Figure 2 for MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models
Figure 3 for MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models
Figure 4 for MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models
Viaarxiv icon