Re:SOTA Code Retrieval with Efficient Code Embedding Models
by timbilt on Monday March 03, 2025 @10:55AM
(#43242945)
anyone else concerned that training models on synthetic, LLM-generated data might push us into a linguistic feedback loop?
relying on LLM text for training could bias the next model towards even more overuse of words like "delve", "showcasing", and "underscores"...