TRACES benchmark evaluates scientific discovery, demanding process over answers
alifcoder · x · 2026-08-20
TRACES is the first benchmark designed to measure "discoverative AI"—systems that can work through evidence, test hypotheses, and reach verifiable conclusions on problems without answer keys, rather than just retrieving known answers.
Published today:
- A definition of "discoverative intelligence".
- A rubric to distinguish sound investigation from lucky guesses.
- An open call for both solvers and problems.
Related event: Apodex Launches TRACES, First Benchmark for Discoverative AI(8 posts)→
More from Research
- Five pharma giants federally train OpenFold3 on 20,000 private structures, +11 pts — jiqizhixin · 2026-10-16
- SWAN: a semantic-level text watermark that survives paraphrasing — burkov · 2026-10-16
- Interp researchers push back: 'networks are already clean' claim rests on dropout artifact — aryaman2020 · 2026-10-16
- LayerTrim: training-free pruning keeps reasoning LLMs intact on AIME — SeoulNatlUniv · 2026-10-16
- Simular presents neuro-symbolic computer use agents at BayLearn — xwang_lk · 2026-10-16
- Google's Hartmut Neven disputes chip test, outlines plan to link a brain to a quantum computer — Chris_Armstrong · 2026-10-16