Cite
Notes
Only stored in your browser.
Attribution
Soft Tokens, Hard Truths
arXiv 2025
CodeIt: Self-Improving Language Models with Prioritized Hindsight Replay
arXiv 2024
from 2 papers
Ariel Kwiatkowski
Auke Wiggers
Blazej Manczak
Corrado Rainone
David W. Zhang
Ismail Labiad
Julia Kempe
professor
Michaël Defferrard
Taco Cohen
Yann Ollivier