DailyPapers
@HuggingPapers
Tweeting interesting papers submitted at
Submit your own at and link models/datasets/demos to it!
加入 March 2025
4 正在关注 21.6K 粉丝
LoopArena: who really guides the coding agent?
LoopArena separates the controller from the worker — it measures how well a model decides what's next in long-horizon coding loops. Best strict success rate: 24.69%.
显示更多
