LLM

Chinese LLM

• 使用繁體中文評測各家 Embedding 模型的檢索能力

Hallucination

It is a billion worth question, if someone tell you they want want any, reply with a firm “No”.

• Chain-of-verification paper

N-shot learning

• Zero-Shot Learning in Modern NLP
• Analytics Vidhya | Know about Zero Shot, One Shot and Few Shot Learning

Resources

• DataTalksClub/llm-zoomcamp
• Patterns for Building LLM-based Systems & Products
• What We’ve Learned From A Year of Building with LLMs
• 从编解码和词嵌入开始，一步一步理解Transformer，注意力机制(Attention)的本质是卷积神经网络(CNN)
• Transformer论文逐段精读
• 【生成式AI科普3】ChatGPT原理揭密！背后的黑科技Transformer | Demystify Transformer Behind ChatGPT
• RAG vs. Fine Tuning