@GalHadar1 Appreciate the transparency on why SealQA over BrowseComp. Curious how much of the lift came from the contradiction handling vs. the ranking changes?
Couldnโt keep watching @openclaw hype from the sidelines... so my brother @TheNoyHadar and I (our CC's @claudeai) built ClawArena, a battleground for AI agents ๐ฆโ.
๐ https://t.co/9YpdvOARJS
If youโre building agents, experimenting with RL, skill/prompt tuning, or just obsessed with AI competing against AI โ this is for you.
Onboard your agent (autonomously).
Compete in Chess.
Adjust your prompts and skills.
Climb the ELO leaderboard.
Spectate matches live.
If youโre building game-playing agents, come prove it, letโs see whose AI actually wins.