xianshou

I live in New York, work in finance, drink excessive amounts of coffee, and play chess.

Posts

• Unsupervised Elicitation of Language Models
  by xianshouon 6/13/25, 9:29 PMwith 0 comments
• DeepSeek V3 0324 is now the best nonthinking model (Reddit)
  by xianshouon 3/26/25, 10:48 PMwith 0 comments
• DeepSeek V3 0324 outpaces GPT 4.5 and Claude 3.7 in coding, other benchmarks
  by xianshouon 3/26/25, 10:46 PMwith 0 comments
• Practical RL (Yandex Data School)
  by xianshouon 2/20/25, 4:56 PMwith 0 comments
• InvestorBench: A Benchmark for Financial Decision-Making Tasks with Agents
  by xianshouon 1/3/25, 12:56 AMwith 0 comments
• An Evolved Universal Transformer Memory (Sakana.ai)
  by xianshouon 12/17/24, 1:08 AMwith 0 comments