Adobe turns eval reference answers into Python functions, lifting LLM-judge MCC from 0.331 to 0.427

dair_ai · x · 2026-09-17

Adobe researchers propose a new approach to agent evaluation: instead of storing a fixed reference answer per eval case — which goes stale when underlying data changes daily — they write each reference answer as a Python function.

Results: agreement with expert labels rises from MCC 0.331 to 0.427, and token cost per case drops 16%. A judge with no ground truth scored MCC -0.379, worse than chance. The pipeline ships as a harness skill; one stated limitation is that the same model, Claude Sonnet 4.6, served as both judge and target.

Original post →

More from coding & agent

coding & agent channel →