Roman Novak

Roman Novak on Hugging Face Daily Papers: 3 papers, 2 in the top 3 of their day, 55 upvotes.

1. Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models 29 upvotes, #2 of 2023-12-12
2. Frontier Language Models are not Robust to Adversarial Arithmetic, or "What do I need to say so you agree 2+2=5? 4 upvotes, #12 of 2023-11-15
3. Small-scale proxies for large-scale Transformer training instabilities 22 upvotes, #3 of 2023-09-26

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.