{"ok":true,"trend":{"id":721792,"platform":"mastodon","region":"global","key":"this is not beautiful, but it's a very useful analysis which exposes some numerical artefacts in my new simulation, whic","title":"This is not beautiful, but it's a very useful analysis which exposes some numerical artefacts in my new simulation, whic","url":"https://mastodon.social/@franco_vazza/117370787850267850","first_seen":"2026-10-02T10:18:08.765615Z","last_seen":"2026-10-02T22:37:28.432810Z","last_rank":1,"peak_rank":1,"last_volume":24,"peak_volume":24,"seen_count":10,"score":0.845234375,"category_hint":"physics","section":"science","category":"physics","summary":"Astrophysicist Franco Vazza says an AI tool helped him uncover numerical artefacts in a new cosmological simulation that he had previously overlooked. He describes the analysis as unattractive but very useful, saying he worked with an LLM trained on a large custom set of instructions. The post is drawing attention for showing AI being used as a genuine debugging aid in scientific research rather than as a novelty.","why":"Researchers and observers are interested in a concrete case of an LLM catching real technical flaws in scientific code.","tone":"neutral","entities":["Franco Vazza","Claude"],"summarized_at":"2026-10-02T22:21:41.728758Z","meta":{"tag":"physics","via":"scan","kind":"status","lang":"en","instance":"mastodon.social","tag_uses":65},"nw":null,"promo":null,"kind":null,"importance":null,"hidden":false,"hide_reason":null,"judged_at":null,"title_en":"Astrophysicist says Claude helped expose artefacts in cosmology simulation","section_name":"Science","category_name":"Physics","timeline":[{"captured_at":"2026-10-02T10:18:08.765615Z","rank":5,"volume":3},{"captured_at":"2026-10-02T11:40:06.658437Z","rank":1,"volume":12},{"captured_at":"2026-10-02T13:02:09.470341Z","rank":1,"volume":15},{"captured_at":"2026-10-02T14:24:28.360253Z","rank":1,"volume":15},{"captured_at":"2026-10-02T15:46:28.189506Z","rank":1,"volume":17},{"captured_at":"2026-10-02T17:08:28.439606Z","rank":1,"volume":17},{"captured_at":"2026-10-02T18:31:09.152734Z","rank":1,"volume":17},{"captured_at":"2026-10-02T19:53:08.603784Z","rank":1,"volume":22},{"captured_at":"2026-10-02T21:15:28.339196Z","rank":1,"volume":23},{"captured_at":"2026-10-02T22:37:28.432810Z","rank":1,"volume":24}],"posts":[{"platform":"mastodon","url":"https://mastodon.social/@franco_vazza/117370787850267850","author":"franco_vazza","title":null,"snippet":"This is not beautiful, but it's a very useful analysis which exposes some numerical artefacts in my new simulation, which I have previously underlooked. For this I used # AI trained an LLM model in Claude using more than 10,000 tokens. Wait, actually for this I trained Larry,…","posted_at":"2026-10-02T10:14:53.911000Z","likes":24}],"elsewhere":[],"window":"7d"}}