evaluation · 4 articles
Featured · recent
OpenAI released 700+ AI math manuscripts, then withdrew three. We split "AI proved X" into five layers, show what a machine checks, and give an audit checklist.
Read →Featured · recent
Reading arXiv:2609.29429 RLCDAlignBench: TypeSafe Jev (RLCD) on 10 alignment failures × 44 benchmarks hits median AUROC 0.886 with a generic question, ~63× cheaper than LLM judges; checklist for wiring Just-Ask-Jev into an agent harness monitor.
Read →Featured · recent
A deep read of arXiv:2609.29769—typed classifier Jev vs three flash LLM judges on nine panels and 5,003 pairs: accuracy seldom separates, cost/latency gaps hit 29–325× and 30–220×, and correlated errors cap a cheap-first cascade at about +1.5pp.
Read →Featured · recent
arXiv:2609.27678 compares Jev and nine hosted/local models on ContractNLI: cost, latency, baseline accuracy, and All-12 persistent correctness rank differently. Close means can hide flips—select on persistent correctness, not accuracy alone.
Read →