LAMDAS: LLM as an Implicit Classifier for Domain-specific Data Selection

Jian Wu, Hang Yu, Bingchang Liu +4

Adapting large language models (LLMs) to specific domains often faces a critical bottleneck: the scarcity of high-quality, human-curated data. While large volumes of unchecked data…