DSAI: Unbiased and Interpretable Latent Feature Extraction for Data-Centric AI

Hyowon Cho, Soonwon Ka, Daechul Park +3

Large language models (LLMs) often struggle to objectively identify latent characteristics in large datasets due to their reliance on pre-trained knowledge rather than actual data…