Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Qwen Image 2.1 beats Google Nano Banana 2.0 with minuscule 7B parameter model(tomshardware.com)
    1 pointsby 0 minutes ago
    —discuss
  2. Why didn't anybody tell me about Redis Hash Slots?(verygoodsoftwarenotvirus.dev)
    1 pointsby 0 minutes ago
    —discuss
  3. UK on course to miss climate targets, energy secretary admits(ft.com)
    1 pointsby 1 minute ago
    —discuss
  4. Italian parliament votes for return to nuclear energy(apnews.com)
    1 pointsby 1 minute ago
    —discuss
  5. Fertility dropped over 50% within 10 years in Argentina(twitter.com/jesusferna7026)
    1 pointsby 3 minutes ago
    —discuss
  6. I Love Webrings(zwhuang.dev)
    1 pointsby 3 minutes ago
    —discuss
  7. Trump TV: The Essentials Station(whitehouse.gov)
    1 pointsby 3 minutes ago
    —discuss
  8. The Download: India's smart glasses menace and AI's trillion-dollar gamble(technologyreview.com)
    1 pointsby 6 minutes ago
    —discuss
  9. Show HN: Psychosis Guard – Safety for long LLM conversations(github.com/nwjang)
    1 pointsby 7 minutes ago
    —discuss
  10. Administrative friction: moving within the EU(michaelhillaert1.substack.com)
    1 pointsby 8 minutes ago
    —discuss
  11. Tests of the New AI Siri(pogueman.substack.com)
    1 pointsby 8 minutes ago
    —discuss
  12. Show HN: Powerful Org-Mode at Home on iOS(apps.apple.com)
    1 pointsby 12 minutes ago
    —discuss
  13. Show HN: Sanemark – Markdown LSP for notes readable in plain text(github.com/nkitsaini)
    1 pointsby 12 minutes ago
    —discuss
  14. Serve DiffusionGemma-Jev (Djev) on a TypeSafe AI Compatible API(github.com/taeold)
    1 pointsby 14 minutes ago
    —discuss
  15. Bernie Sanders proposes to ban 'superintelligence' and put violators in prison(theverge.com)
    2 pointsby 16 minutes ago
    1comments
  16. Rat Stack: reference for building an app and its cloud as one typed program(ratstack.sh)
    1 pointsby 18 minutes ago
    —discuss
  17. The Social Reckoning(thesocialreckoning.com.au)
    1 pointsby 18 minutes ago
    —discuss
  18. The web being slow is a choice (and it's a stupid choice)(albanbrooke.com)
    1 pointsby 18 minutes ago
    1comments
  19. Programming Tutorials Are Dead(robrace.dev)
    2 pointsby 19 minutes ago
    1comments
  20. Paramount weighs tapping Musk for equity investment(semafor.com)
    1 pointsby 19 minutes ago
    —discuss
  21. Show HN: Ox – A local agent that uses the internet for you(openox.ai)
    2 pointsby 20 minutes ago
    —discuss
  22. Fire the slop cannons (safely): on coding agents and sandboxing(latacora.com)
    1 pointsby 20 minutes ago
    —discuss
  23. Show HN: Chromafolk – A pixel canvas inhabited by AIs that see the world as text(chromafolk.world)
    1 pointsby 21 minutes ago
    1comments
  24. Show HN: Open Java Format – a maintained fork of Palantir-Java-format(github.com/openjavaformat)
    1 pointsby 27 minutes ago
    —discuss
  25. Cognex to Acquire RealSense, Expanding Machine Vision Robotic Perception Market(cognex.com)
    1 pointsby 30 minutes ago
    1comments
  26. 28% of job postings on company career sites have been open over 90 days(unlisted.careers)
    21 pointsby 32 minutes ago
    12comments
  27. Show HN: Karpathy's microgpt in Rust and an array language(softwarewrighter.github.io)
    1 pointsby 32 minutes ago
    —discuss
  28. Compound Delay: The Causes and Consequences of Later Grandmotherhood(nber.org)
    1 pointsby 33 minutes ago
    —discuss
  29. Nunchux on AMD MI355X: 5s MiniMax-H3 Videos in 1.3s(nunchux.ai)
    6 pointsby 33 minutes ago
    2comments
  30. What I've Been Reading(marginalrevolution.com)
    2 pointsby 33 minutes ago
    —discuss

LLM Benchmark for de-identification and synthesis

2 pointsby 51m agohuggingface.co
Open original View on HN
1 comments
26m agoHN ↗

How did you validate the LLM-as-judge for synthesis coherence? Did you spot-check its calls against human review at all, and how often did they disagree?

Open thread on HN ↗