1 citations · 1 across the 4 of their papers we have counts for
1 paper · 1 filter
Animesh Jha, Harshit Gupta, Ananjan Nandi
Data selection for finetuning Large Language Models (LLMs) can be framed as a budget-constrained optimization problem: maximizing a model's downstream performance under a strict tr…