Middo: Model-Informed Dynamic Data Optimization for Enhanced LLM Fine-Tuning via Closed-Loop Learning
Paper
•
2508.21589
•
Published
•
3
Middo
Dataset & Models for paper "Middo: Model-Informed Dynamic Data Optimization for Enhanced LLM Fine-Tuning via Closed-Loop Learning"
-
-
Word2Li/MiddOptimized
Viewer • Updated • 411k • 228 -
Word2Li/MiddOnly
Viewer • Updated • 33.7k • 139 -
Word2Li/Llama3.1-8B-Middo-Alpaca
Text Generation • 8B • Updated • 45 -
Word2Li/Llama3.1-8B-Middo-Alpaca-4o-mini
Text Generation • 8B • Updated • 23 -
Word2Li/Llama3.1-8B-Middo-Wizard
Text Generation • 8B • Updated • 19 • 1 -
Word2Li/Mistral-7B-v0.3-Middo-Alpaca
Text Generation • 7B • Updated • 16 -
Word2Li/Mistral-7B-v0.3-Middo-Alpaca-4o-mini
Text Generation • 7B • Updated • 24 • 1 -
Word2Li/Mistral-7B-v0.3-Middo-Wizard
Text Generation • 7B • Updated • 15 • 1 -
Word2Li/Alpaca-4o-mini
Viewer • Updated • 52k • 91