Yu-Ting Lee (Quert)

這篇論文探討了知識密集型問答（QA）任務中，如何利用規則來增強現有檢索增強生成（RAG）框架的效能。

Fine-tuning Embeddings with NUDGE

Fine-tuning Gemma 2 9B with Unsloth

Fine-tuning Gemma 2 9B with PEFT and LoRA

What Matters in Transformers? Not All Attention is Needed

Advanced RAG with Knowledge Graphs: Using neo4j to implementate RAG with KG

Research Paper: NUDGE: Lightweight Non-Parametric Fine-Tuning of Embeddings for Retrieval

Stima Research 是一個 Stima Tech 所屬的 NLP 研究團隊，致力於探索 AI 的前沿技術。