AI RESEARCH
Explore Before You Solve: The Speed--Depth Trade-off in Epistemic Agents for ARC-AGI-3
arXiv CS.AI
•
ArXi:2605.25931v1 Announce Type: new We systematically investigate all 25 public ARC-AGI-3 games and find that every one is reachable through non-intelligent strategies: 10 in a single blind step, 5 after one probing action, 1 via repeated ACTION1 presses, 1 via diverse exploration, and 8 via single repeated actions with sufficient budget (50-200 steps). A library-level null-coordinate vulnerability additionally bypasses 18 games in 1 step.