The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation Paper • 2609.36484 • Published 8 days ago • 578
interlinguistic-language-modeling/BabyLM-130M-ilm_it-ilm_pl-tokenizer_ilm_bilingual_it_pl_bpe_trailing_ws_32000 Text Generation • 0.1B • Updated Apr 21 • 356 • 3