2 papers
cs.AI2026
How Much of a Harness Does a Strong Agent Need for Autonomous ML Engineering?
Kirill Brilliantov, Alejandro Hernández-Cano, Emmanuel Abbé
Recent autonomous machine learning engineering (MLE) agents have made significant progress on public leaderboards. Often motivated by progress stagnation over long-horizon cycles a…
cs.MS2026
-server-bench: Automating Potential Discovery for the -Server Conjecture
Kirill Brilliantov, Etienne Bamas, Emmanuel Abbé
We introduce a code-based challenge for automated, open-ended mathematical discovery based on the -server conjecture, a central open problem in competitive analysis. The task is…