Vals AI's MysteryMechanism Benchmark Tests If AI Can Rediscover 222 Scientific Laws
A new benchmark seals 222 scientific laws and asks agents to rediscover each one from scratch using a tight experiment budget, with GPT-6 Astra leading at 53.2%.
Source
Vals AI
Published
Author
AlphaSignal Newsroom
Read
1 min read
A new benchmark seals 222 scientific laws and asks agents to rediscover each one from scratch using a tight experiment budget, with GPT-6 Astra leading at 53.2%.
Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.
