| 1. | | An OpenAI model has disproved a central conjecture in discrete geometry (openai.com) |
| 1429 points by tedsanders 13 days ago | 1054 comments |
|
| 2. | | Talkie: a 13B vintage language model from 1930 (talkie-lm.com) |
| 773 points by jekude 36 days ago | 331 comments |
|
| 3. | | All elementary functions from a single binary operator (arxiv.org) |
| 858 points by pizza 50 days ago | 298 comments |
|
| 4. | | Missile defense is NP-complete (smu160.github.io) |
| 382 points by O3marchnative 70 days ago | 427 comments |
|
| 5. | | Executing programs inside transformers with exponentially faster inference (percepta.ai) |
| 330 points by u1hcw9nx 82 days ago | 130 comments |
|
| 6. | | Show HN: Steerling-8B, a language model that can explain any token it generates (guidelabs.ai) |
| 328 points by adebayoj 3 months ago | 91 comments |
|
| 7. | | The Waymo World Model (waymo.com) |
| 1160 points by xnx 3 months ago | 663 comments |
|
| 8. | | Which AI Lies Best? A game theory classic designed by John Nash (so-long-sucker.vercel.app) |
| 195 points by lout332 4 months ago | 80 comments |
|
| 9. | | Dynamic Large Concept Models: Latent Reasoning in an Adaptive Semantic Space (arxiv.org) |
| 55 points by gmays 4 months ago | 5 comments |
|
| 10. | | Universal Reasoning Model (53.8% pass 1 ARC1 and 16.0% ARC 2) (arxiv.org) |
| 131 points by marojejian 5 months ago | 30 comments |
|
| 11. | | Auto-grading decade-old Hacker News discussions with hindsight (karpathy.bearblog.dev) |
| 686 points by __rito__ 5 months ago | 270 comments |
|
| 12. | | Show HN: WhatHappened – HN summaries, heatmaps, and contrarian picks (whathappened.tech) |
| 66 points by marsw42 5 months ago | 36 comments |
|
| 13. | | Terrain Diffusion: A Diffusion-Based Successor to Perlin Noise (arxiv.org) |
| 154 points by kelseyfrog 5 months ago | 42 comments |
|
| 14. | | The universal weight subspace hypothesis (arxiv.org) |
| 358 points by lukeplato 5 months ago | 132 comments |
|
| 15. | | In Soviet Union, Optimization Problem Solves You (2012) (crookedtimber.org) |
| 92 points by mpweiher on Nov 3, 2018 | 109 comments |
|
| 16. | | Want to piss off your IT department? Are the links not malicious looking enough? (phishyurl.com) |
| 1127 points by jordigh 8 months ago | 320 comments |
|
| 17. | | Why language models hallucinate (openai.com) |
| 278 points by simianwords 8 months ago | 224 comments |
|
| 18. | | Reinforcement learning is supervised learning on optimized data (bair.berkeley.edu) |
| 241 points by jonbaer on Oct 14, 2020 | 18 comments |
|
| 19. | | Show HN: 0xDEAD//TYPE – A fast-paced typing shooter with retro vibes (theden.sh) |
| 117 points by theden 10 months ago | 28 comments |
|
| 20. | | NeuralOS: An operating system powered by neural networks (neural-os.com) |
| 208 points by yuntian 10 months ago | 55 comments |
|
| 21. | | Compiling LLMs into a MegaKernel: A path to low-latency inference (zhihaojia.medium.com) |
| 314 points by matt_d 11 months ago | 76 comments |
|
| 22. | | Q-learning is not yet scalable (seohong.me) |
| 220 points by jxmorris12 11 months ago | 48 comments |
|
| 23. | | Differences in link hallucination and source comprehension across different LLM (mikecaulfield.substack.com) |
| 78 points by hveksr 12 months ago | 42 comments |
|
| 24. | | Transformer neural net learns to run Conway's Game of Life just from examples (sidsite.com) |
| 69 points by montebicyclelo on May 17, 2025 | 35 comments |
|
| 25. | | AlphaEvolve: A Gemini-powered coding agent for designing advanced algorithms (deepmind.google) |
| 1036 points by Fysi on May 14, 2025 | 270 comments |
|
| 26. | | Mipmap selection in too much detail (pema.dev) |
| 99 points by luu on May 11, 2025 | 25 comments |
|
| 27. | | Type-constrained code generation with language models (arxiv.org) |
| 257 points by tough on May 13, 2025 | 127 comments |
|
| 28. | | Benchmarking LLM social skills with an elimination game (github.com/lechmazur) |
| 194 points by colonCapitalDee on April 4, 2025 | 60 comments |
|
| 29. | | AI 2027 (ai-2027.com) |
| 949 points by Tenoke on April 3, 2025 | 621 comments |
|
| 30. | | James Webb Space Telescope reveals that most galaxies rotate clockwise (smithsonianmag.com) |
| 356 points by instagraham on March 31, 2025 | 310 comments |
|
|
| More |
