Peml

PEML method optimizes LLM prompts and weights for multi-task learning

Researchers have introduced PEML, a new method for parameter-efficient multi-task learning in large language models. PEML optimizes both continuous prompts and model weights simultaneously, addressing limitations of exi…