[arteval-bench] Integrate ae-agent into ArtEval benchmark - #131

Merged

xuafeng merged 7 commits into

sys-intelligence:mainfrom

Couen:feature/ae-agent-arteval-bench

Mar 5, 2026

Commits on Feb 27, 2026

chore: unify task jsonl format - use env instead of docker_env/docer_env, omit optional fields
Couen
committed
cefd7c1

Commits on Mar 1, 2026

Refactor arteval bench: ae_agent integration and code cleanup
Couen
committed
511bfa3
Fix evaluator invocation: run Python scripts with python from repo root
Couen
committed
d9af6f1