Decoding as Optimisation on the Probability Simplex: From Top-K to Top-P (Nucleus) to Best-of-K Samplers Paper โข 2602.18292 โข Published 26 days ago โข 10 โข 6
Decoding as Optimisation on the Probability Simplex: From Top-K to Top-P (Nucleus) to Best-of-K Samplers Paper โข 2602.18292 โข Published 26 days ago โข 10
Learning Query-Specific Rubrics from Human Preferences for DeepResearch Report Generation Paper โข 2602.03619 โข Published Feb 3 โข 27
view post Post 3052 Just built my entire AI Engineer portfolio by pasting 2 links (GitHub and LinkedIn) into moonshotai Kimi 2.5.That's it. That's the workflow.Zero coding. Zero iteration. Zero "make the button bigger."See for yourself: https://rakshit2020.github.io/rakshitaralimatti.github.io/The model:โ Scraped my GitHub repos automaticallyโ Pulled my experience from LinkedInโ Designed an Aurora Glass themeโ Mapped every skill to projectsโ Added animations I'd never code myself See translation 4 replies ยท ๐ 10 10 ๐ 4 4 + Reply
Post-LayerNorm Is Back: Stable, ExpressivE, and Deep Paper โข 2601.19895 โข Published Jan 27 โข 25
view article Article Unlocking Agentic RL Training for GPT-OSS: A Practical Retrospective Jan 27 โข 67
Running 39 Korean Open Source Heatmap ๐ฅ 39 Explore Korean AI open source contributions and releases
Over-Searching in Search-Augmented Large Language Models Paper โข 2601.05503 โข Published Jan 9 โข 7