ForeSci Benchmark

What is ForeSci?

Temporally controlled benchmark with 500 tasks across AI domains for forward-looking research judgment; includes offline knowledge bases and validation…

Released

2026-05-30

Evaluates

General AI, Language & Knowledge, cs.AI

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2606.00644

ForeSci paper

• Paper https://arxiv.org/abs/2606.00644

Which sources cite ForeSci?

• ForeSci

Related General AI, Language & Knowledge, cs.AI benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A Large-Scale Dataset and Benchmark
• A3S-Bench
• AARRI-Bench
• ABC-Bench
• ActBench