Hackernews posts about DEC

1. 961
   

   newI built non-autoregressive decision models with RL a year agolaya.convaiinnovations.com

   ▲ 1363nandakishor_ml14d💬 319savesaved
2. 962
   

   newLibreOffice breaks download records after declaring it has no AI featuresmanualdousuario.net

   ▲ 722rpgbr25d💬 237savesaved
3. 963
   

   newClef: Open-weight decision models, and new RL fine-tuning platformblog.cloudflare.com

   ▲ 623jasondavies2d💬 215savesaved
4. 964
   

   newOllaya – Ollama for open-source, Jev-style decision modelsollaya.dev

   ▲ 617Ardakilic8d💬 145savesaved
5. 965
   

   newJeff – Jev-compatible 0.8B decision models, trained at home, ~30 msgithub.com

   ▲ 574firelex4d💬 225savesaved
6. 966
   

   newKev: Tiny Jev-like family of decision models built on top of Qwen3.5github.com

   ▲ 462tosh12d💬 211savesaved
7. 967
   

   newFable 5 – Median thinking declined in Augusttwitter.com

   ▲ 427espeed12d💬 293savesaved
8. 968
   

   newJury finds Facebook liable for deceiving users in Cambridge Analytica casewww.cbsnews.com

   ▲ 423pseudolus7d💬 107savesaved
9. 969
   

   newShow HN: Reladraw – A diagram language where you decide where to place thingsgithub.com

   ▲ 411jpwalsh2347d💬 119savesaved
10. 970
    

    newParley: Federated, decentralised chat that speaks plain IRCgit.mills.io

    ▲ 328davidcollantes5d💬 190savesaved
11. 971
    

    newJeeves. Reasoning improves Jev-like decision modelsgithub.com

    ▲ 242nicowaltz4d💬 95savesaved
12. 972
    

    newUsing LLMs to trace alchemical knowledge and decode 17th century lettersresobscura.substack.com

    ▲ 174benbreen8d💬 44savesaved
13. 973
    

    newThe most important product decision is what you don't buildliamnugent.me

    ▲ 161ChrisArchitect15d💬 54savesaved
14. 974
    

    newShow HN: JevBench, a reproducible benchmark for typed decision modelsbenchmarkheaven.com

    ▲ 153florianstandhar11d💬 39savesaved
15. 975
    

    newDecoding the NEC V20 Microcodemartypc.blogspot.com

    ▲ 150mariuz29d💬 10savesaved
16. 976
    

    newSpeculative Decoding in vLLM on AMD GPUsvllm.ai

    ▲ 145ankitg1226d💬 53savesaved
17. 977
    

    newResident Evil 4 (GameCube) – complete byte-identical decompilation to C/C++github.com

    ▲ 140metrofun13d💬 99savesaved
18. 978
    

    newTurning GLM-5.3-Flash into a Jev-like decision modelwww.privatemode.ai

    ▲ 138flxflx7d💬 59savesaved
19. 979
    

    newWe've followed their lives for six decades; now the stars of 7 Up are bowing outwww.bbc.co.uk

    ▲ 132mellosouls21d💬 37savesaved
20. 980
    

    newPISA 2025 Students' reading and mathematics performance declined across the OECDwww.oecd.org

    ▲ 131mazokum25d💬 169savesaved
21. 981
    

    newTelstra outage: The night a network decided the year was 2006www.netnod.se

    ▲ 93TMWNN15d💬 34savesaved
22. 982
    

    newShow HN: Air-gapped file encryption as self-decrypting HTML pagecms-sfx-demo.apeleg.com

    ▲ 90emurlin9d💬 32savesaved
23. 983
    

    newF.02 Decommissionwww.figure.ai

    ▲ 86ad_hockey32h💬 42savesaved
24. 984
    

    newDecapitating a MacBook (2025)mm-dev.rocks

    ▲ 67luu27d💬 48savesaved
25. 985
    

    newAdding Floating-Point Decimals for Fun and Profitblog.vero.site

    ▲ 62ibobev4d💬 35savesaved
26. 986
    

    newDeCloudflare0xacab.org

    ▲ 52DeepLogin21d💬 55savesaved
27. 987
    

    newShow HN: jevals – replacing LLM judges with typed Jev decisionsgithub.com

    ▲ 47gbayomi12d💬 6savesaved
28. 988
    

    newShow HN: AI·rete·RAG – a Rete rule engine decides, RAG explains whyai-rete-rag.com

    ▲ 46ZaharaHussain11d💬 11savesaved
29. 989
    

    newWhen anyone can build software, who decides what not to build?architectureintel.com

    ▲ 44younss20d💬 35savesaved
30. 990
    

    newThe great crime decline is happening all across the countrywww.theatlantic.com

    ▲ 42andsoitis5d💬 64savesaved

· edit