Middo: Model-Informed Dynamic Data Optimization for Enhanced LLM Fine-Tuning via Closed-Loop Learning (EMNLP 2025 Main)
-
Updated
May 18, 2026 - Python
Middo: Model-Informed Dynamic Data Optimization for Enhanced LLM Fine-Tuning via Closed-Loop Learning (EMNLP 2025 Main)
Code base for actively eliciting expensive/ cumbersome features of the most informative data instance. (IJCAI 2018)
To associate your repository with the dataselection topic, visit your repo's landing page and select "manage topics."