Jeremy's excellent work here is a great illustration of a very powerful type of approach: LLM-guided on-the-fly synthesis of a symbolic world model, i.e. making sense of the world by writing executable code that encodes your understanding of the causal mechanics of the world.
So far, all of the top-performing harnesses on ARC-AGI-3 use this style of approach. Which is also the approach we recommended when we initially released the benchmark (Jeremy would know this better than most, as a former Ndea member of technical staff).
I'm happy that ARC 3 has incentivized more research and more progress in this area.
In one pass Opus wrote 269 programs (~12,700 lines). It built parsers for all 25 games, searching functions for 23, and game simulators for 9. It built a differ...