{"ok":true,"trend":{"id":806975,"platform":"news","region":"global","key":"new mit and sakana ai framework uses an llm judge to cut evaluation costs for self-improving coding agents","title":"New MIT and Sakana AI framework uses an LLM judge to cut evaluation costs for self-improving coding agents","url":"https://news.google.com/rss/articles/CBMi3AFBVV95cUxNSDVzdEplY29renRTNm9PR1Z5QjFpak9tb2RQc2FlQzV0YVlGVDdIZmEtZE9ia28wMmFpakthckd4ejlHN0FpTnBwZHY5ZUxVTW52NTg0ajR1VkpMTExGOHZKdV90VFRqV3p1UTFJSTQ2LWRxWjRScE9FUE1MVHcwcHV2TGlRdmJCR1J5VFlncXMyRVhnVGRwa3dxaDBpUkRpV3dLV3F4T19aTzVYblNNZXdlSmZzTmNTREhSem81TXNYNmNJZHdhT3JDT0pQMmxzZWd6M1NuTFVQb2lf?oc=5","first_seen":"2026-10-03T02:35:12.414991Z","last_seen":"2026-10-03T02:35:12.414991Z","last_rank":37,"peak_rank":37,"last_volume":null,"peak_volume":null,"seen_count":1,"score":0.5478125,"category_hint":"ai","section":"technology","category":"ai","summary":"MIT and Sakana AI have introduced a new framework that uses a large language model as an automated judge to evaluate the output of self-improving coding agents. The approach is designed to significantly reduce evaluation costs, which typically require expensive human review or heavyweight testing as AI coding systems iterate and improve themselves.","why":"Cutting evaluation costs addresses a major bottleneck as AI labs race to build coding agents that improve themselves.","tone":"neutral","entities":["MIT","Sakana AI"],"summarized_at":"2026-10-03T02:37:39.991811Z","meta":{"via":"scan","lang":"en","term":"llm","source":"VentureBeat","published":"Fri, 02 Oct 2026 22:50:08 GMT"},"nw":null,"promo":null,"kind":null,"importance":null,"hidden":false,"hide_reason":null,"judged_at":null,"title_en":"MIT and Sakana AI unveil cheaper evaluation for self-improving coding agents","section_name":"Technology","category_name":"AI","timeline":[{"captured_at":"2026-10-03T02:35:12.414991Z","rank":37,"volume":null}],"posts":[{"platform":"news","url":"https://news.google.com/rss/articles/CBMi3AFBVV95cUxNSDVzdEplY29renRTNm9PR1Z5QjFpak9tb2RQc2FlQzV0YVlGVDdIZmEtZE9ia28wMmFpakthckd4ejlHN0FpTnBwZHY5ZUxVTW52NTg0ajR1VkpMTExGOHZKdV90VFRqV3p1UTFJSTQ2LWRxWjRScE9FUE1MVHcwcHV2TGlRdmJCR1J5VFlncXMyRVhnVGRwa3dxaDBpUkRpV3dLV3F4T19aTzVYblNNZXdlSmZzTmNTREhSem81TXNYNmNJZHdhT3JDT0pQMmxzZWd6M1NuTFVQb2lf?oc=5","author":"VentureBeat","title":"New MIT and Sakana AI framework uses an LLM judge to cut evaluation costs for self-improving coding agents","snippet":null,"posted_at":"2026-10-02T22:50:08Z","likes":null}],"elsewhere":[],"window":"7d"}}