#evaluation

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.

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.