Y Hacker News new | ask | show | jobs

user: xianshou created: 2011-05-01 karma: 2813 I live in New York, work in finance, drink excessive amounts of coffee, and play chess. submissions: 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments Merge and Conquer: Evolutionarily Optimizing AI for 2048 1 points | 0 comments Stuck in the Matrix: Probing Spatial Reasoning in Large Language Models 1 points | 0 comments Reflection AI Raises $2B to Build "American DeepSeek" 9 points | 2 comments Nvidia-backed Reflection AI raising at $5.5B valuation 2 points | 1 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments Unsupervised Elicitation of Language Models 7 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments DeepSeek V3 0324 is now the best nonthinking model (Reddit) 1 points | 0 comments DeepSeek V3 0324 outpaces GPT 4.5 and Claude 3.7 in coding, other benchmarks 7 points | 0 comments 0 points | 0 comments Practical RL (Yandex Data School) 1 points | 0 comments 0 points | 0 comments