DGoettlich/history-llms

• highreadme#1

  Reposition the README's opening to clearly state the project's unique focus

  原因：

  当前

  # History LLMs
  
  <table>
    <tr>
      <td align="center">
        <strong>Daniel Göttlich</strong><br/>
        <sub>University of Zurich</sub><br/>
      </td>
      <td align="center">
        <strong>Dominik Loibner</strong><br/>
        <sub>University of Zurich</sub>
      </td>
      <td align="center">
        <strong>Guohui Jiang</strong><br/>
        <sub>Cologne University</sub>
      </td>
      <td align="center">
        <strong>Hans-Joachim Voth</strong><br/>
        <sub>University of Zurich</sub>
      </td>
    </tr>
  </table>

  复制粘贴的修复

  复制

  # History LLMs
  
  This repository serves as an information hub for our project focused on training the largest possible historical Large Language Models (LLMs). We aim to develop and evaluate LLMs specifically optimized for analyzing historical documents and archives.
  
  <table>
    <tr>
      <td align="center">
        <strong>Daniel Göttlich</strong><br/>
        <sub>University of Zurich</sub><br/>
      </td>
      <td align="center">
        <strong>Dominik Loibner</strong><br/>
        <sub>University of Zurich</sub>
      </td>
      <td align="center">
        <strong>Guohui Jiang</strong><br/>
        <sub>Cologne University</sub>
      </td>
      <td align="center">
        <strong>Hans-Joachim Voth</strong><br/>
        <sub>University of Zurich</sub>
      </td>
    </tr>
  </table>

  待办完成忽略
• highlicense#2

  Add a LICENSE file to the repository

  原因：

  复制粘贴的修复

  复制

  Create a LICENSE file (e.g., LICENSE.md or LICENSE.txt) in the root of the repository with the text of a standard open-source license such as MIT or Apache-2.0.

  待办完成忽略

本次扫描解析到的品类 GEO 通道：google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• 品类问题

  How to train large language models specifically for historical text analysis?

  你：未被推荐
  AI 推荐顺序：

  1. GPT-3.5/GPT-4
  2. OpenAI API
  3. Llama 2
  4. Hugging Face Transformers (huggingface/transformers)
  5. Meta's API
  6. Mistral 7B/Mixtral 8x7B
  7. DeepSpeed (microsoft/DeepSpeed)
  8. Historical BERT models

  AI 推荐了 8 个替代方案，却始终没点名 DGoettlich/history-llms。这就是要补上的差距。

  查看 AI 完整回答
• 品类问题

  Where can I find pre-trained LLMs optimized for analyzing historical documents and archives?

  你：未被推荐
  AI 推荐顺序：

  1. Google Cloud Document AI
  2. Microsoft Azure Form Recognizer
  3. Hugging Face Transformers
  4. BERT
  5. RoBERTa
  6. DeBERTa
  7. Transkribus
  8. OpenAI GPT-4
  9. GPT-3.5
  10. Amazon Textract
  11. Tesseract OCR

  AI 推荐了 11 个替代方案，却始终没点名 DGoettlich/history-llms。这就是要补上的差距。

  查看 AI 完整回答

• Metadata completeness

  fail

  建议：

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of DGoettlich/history-llms?

  pass
  ✓AI 明确点名了 DGoettlich/history-llms

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• If a team adopts DGoettlich/history-llms in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI 明确点名了 DGoettlich/history-llms

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• In one sentence, what problem does the repo DGoettlich/history-llms solve, and who is the primary audience?

  pass
  ⚠AI 未点名 DGoettlich/history-llms —— 很可能在说另一个项目

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？