TRACES: First Benchmark to Measure 'Discoverative' AI in Science
hey_abusiddik · x · 2026-08-22
Apodex introduced TRACES, the world's first benchmark designed to measure 'discoverative AI.' Unlike traditional retrieval benchmarks that test finding known answers, TRACES evaluates the entire research process—tool use, evidence gathering, self-correction, and testing alternative hypotheses—to solve problems without answer keys. The project defines 'discoverative intelligence' and provides a rubric to distinguish sound investigation from lucky guesses, currently open for solvers and problems.
Related event: Apodex Launches TRACES Benchmark for AI Scientific Discovery(2 posts)→
More from AGI Musings
- False Collatz Proof via a Lean Bug: Talk Probes Proof Assistants as LLM-Found Bug Targets — marc_lelarge · 2026-10-16
- Pennsylvania residents reject $10,000-per-household offer for massive Amazon data center — Polymarket · 2026-10-16
- A 1968 newspaper predicted today's AI art debate almost word for word — mtizard · 2026-10-16
- Curing AI-era FOMO: define what 'fulfilling your potential' means to you — dejavucoder · 2026-10-16
- "An LLM doesn't understand anything" gets the perfect parody reversal: neither do you — VoidStateKate · 2026-10-16
- Nina Schick: Europe's Policy Circle Still Doesn't Believe FSD, Orbital Compute Are Real — NinaDSchick · 2026-10-16