Picture for Zhengyan Shi

Zhengyan Shi

When Can Proxies Improve the Sample Complexity of Preference Learning?

Add code
Dec 21, 2024
Viaarxiv icon

Understanding Likelihood Over-optimisation in Direct Alignment Algorithms

Add code
Oct 15, 2024
Figure 1 for Understanding Likelihood Over-optimisation in Direct Alignment Algorithms
Figure 2 for Understanding Likelihood Over-optimisation in Direct Alignment Algorithms
Figure 3 for Understanding Likelihood Over-optimisation in Direct Alignment Algorithms
Figure 4 for Understanding Likelihood Over-optimisation in Direct Alignment Algorithms
Viaarxiv icon

Understanding the Role of User Profile in the Personalization of Large Language Models

Add code
Jun 22, 2024
Figure 1 for Understanding the Role of User Profile in the Personalization of Large Language Models
Figure 2 for Understanding the Role of User Profile in the Personalization of Large Language Models
Figure 3 for Understanding the Role of User Profile in the Personalization of Large Language Models
Figure 4 for Understanding the Role of User Profile in the Personalization of Large Language Models
Viaarxiv icon

Instruction Tuning With Loss Over Instructions

Add code
May 23, 2024
Viaarxiv icon