• How might LLMs store facts | DL7
  • Refer to: https://x.com/howie_serious/status/1889471949136601109
  • https://www.youtube.com/watch?v=9-Jl0dxWQs8
• 最好的致敬是学习：DeepSeek-R1 赏析
  • https://mp.weixin.qq.com/s/_XGBipbywCOtcKu13QDW5Q
• Understanding Reasoning LLMs ｜ 理解推理型大语言模型
  • https://magazine.sebastianraschka.com/p/understanding-reasoning-llms
  • https://mp.weixin.qq.com/s/DG8-bENYNji8qg4SwexEmg
• Deep Dive into LLMs like ChatGPT
  • https://x.com/karpathy/status/1887211193099825254
  • https://www.youtube.com/watch?v=7xTGNNLPyMI

https://x.com/dotey/status/1888095733154201674

• On the consistent reasoning paradox of intelligence and optimal trust in AI: The power of ‘I don’t know’
  • https://arxiv.org/html/2408.02357v1

模型能力测评

• FlagEval 天秤大模型评测平台：https://flageval.baai.ac.cn/

https://x.com/karminski3/status/1912556273897468415

https://openai.com/index/gpt-4-1/

https://openai.com/index/introducing-o3-and-o4-mini/