4,407 Matching Annotations
  1. Jul 2025
  2. file.notion.so file.notion.so
    wellsenn XR拆解及BOM成本报告：逸文Even G1 AR眼镜 - 免费版.pdf
    1
    Visit annotations in context

    Annotators

    URL

    file.notion.so/f/f/e1806fae-3fdb-4c82-9df1-31a050cbbd4a/531d2ee5-d043-4d18-804b-3f73e27543b5/wellsenn_XR拆解及BOM成本报告：逸文Even_G1_AR眼镜_-_免费版.pdf
  3. moonshotai.github.io moonshotai.github.io
    Kimi K2：开放自主智能 --- Kimi K2: Open Agentic Intelligence
    1
    Visit annotations in context

    Annotators

    URL

    moonshotai.github.io/Kimi-K2/
  4. x.com x.com
    (10) Rainman on X: "评估方法： 本次评估主要考察模型在以下四项任务中的表现： 单跳任务评估（Single Hop）：测试模型在已知上下文中对单一事实的直接回忆能力。 多跳任务评估（Multi Hop）：考察模型能否通过多轮推理整合分散信息。 开放问题评估（Open Domain）：评估模型在非限定问题上的记忆准确性和灵活性。 https://t.co/peznVBSai9" / X
    1
    Visit annotations in context

    Annotators

    URL

    x.com/0xdeusyu/status/1942089566690984431
  5. Mar 2025
  6. docs.nvidia.com docs.nvidia.com
    Parallelisms — NVIDIA NeMo Framework User Guide
    1
    Visit annotations in context

    Annotators

    URL

    docs.nvidia.com/nemo-framework/user-guide/24.09/nemotoolkit/features/parallelisms.html
  7. Nov 2024
  8. githubnext.com githubnext.com
    GitHub Next | GitHub Spark
    2
    Visit annotations in context

    Annotators

    URL

    githubnext.com/projects/github-spark
  9. Sep 2024
  10. openai.com openai.com
    Learning to Reason with LLMs
    1
    1. fxp007 19 Sep 2024
      consistently improves with more reinforcement learning (train-time compute) and with more time spent thinking (test-time compute)

      RL for post-train, time spent thinking for inference? How?

    Visit annotations in context

    Annotators

    URL

    openai.com/index/learning-to-reason-with-llms/