Picture for Jinhong Yu

Jinhong Yu

ROSE: A Reward-Oriented Data Selection Framework for LLM Task-Specific Instruction Tuning

Add code
Dec 01, 2024
Figure 1 for ROSE: A Reward-Oriented Data Selection Framework for LLM Task-Specific Instruction Tuning
Figure 2 for ROSE: A Reward-Oriented Data Selection Framework for LLM Task-Specific Instruction Tuning
Figure 3 for ROSE: A Reward-Oriented Data Selection Framework for LLM Task-Specific Instruction Tuning
Figure 4 for ROSE: A Reward-Oriented Data Selection Framework for LLM Task-Specific Instruction Tuning
Viaarxiv icon