Home/Blog/BenchAudit – Auditing LLM benchmarks for leaks, data reuse, and harness gapsAI & MLBenchAudit – Auditing LLM benchmarks for leaks, data reuse, and harness gapsHacker News New2026-11-18 15:410 viewsArticle URL: https://benchaudit.org/ Comments URL: https://news.ycombinator.com/item?id=50543033 Points: 1 # Comments: 0Source:https://benchaudit.org/My favoritesPreviousOriko CLI – Create your CV in the terminal with builtin Agent supportNextCodeberg seems to have vanished from DNSRelated postsGuillotine DJ – Announcement TrailerArticle URL: https://www.youtube.com/watch?v=pma19x8XSFs Comments URL: https://news.ycombinator.com/item?id=50543061 Poi0Hacker News NewResearchers hijack Hugging Face account for popular Kokoro AI modelArticle URL: https://www.bleepingcomputer.com/news/security/researchers-hijack-hugging-face-account-for-popular-kokoro-a0Hacker News NewAutomated LLM Fine-Tuning: GPU Selection, Dataset Analysis and FeedbackArticle URL: https://sigvard.ai Comments URL: https://news.ycombinator.com/item?id=50543212 Points: 1 # Comments: 00Hacker News NewHuman-written Launch HN posts get more upvotes than AI-generated onesArticle URL: https://www.lettercamp.com/blog/ai-writing-in-launch-hn-posts Comments URL: https://news.ycombinator.com/it0Hacker News NewParallels Desktop reviewApple offers some of the best workstations going, with everything from the MacBook Pro to Mac mini finding its die-hard 0TechRadarHow Far Behind Are Open Weights, MeasuredWhere AI Research Actually Happens ended on one bar chart: the best open-weight models trail the closed frontier by four0dev.to