CategoryAgentic
MeasureHeld-out accuracy (Best@3)
Tasks70 held-out program-synthesis tasks; 31 discovery targets
DifficultyDiscovering and applying unfamiliar executable rules
Tests whether a system can discover a programming language's hidden rules through experiments and apply them to unseen tasks.
No model in the catalog has a published ExplorationBench AlienCode score.The benchmark is defined by ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien Worlds, but the catalog carries no value for it yet. An absent value is shown as absent here rather than as a zero.
About ExplorationBench AlienCode
What the benchmark measures and how its value is published, from the source’s own record.
What is ExplorationBench AlienCode?
ExplorationBench AlienCode (ExplorationBench AlienCode Best@3) is an agentic benchmark published by ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien Worlds. Tests whether a system can discover a programming language's hidden rules through experiments and apply them to unseen tasks. Its evaluation set is described by the source as 70 held-out program-synthesis tasks; 31 discovery targets. The source rates its difficulty as discovering and applying unfamiliar executable rules.
How is ExplorationBench AlienCode scored?
Each model's result is published by ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien Worlds as held-out accuracy (Best@3). ModelScale prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 0 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
ModelScale