Open LLM Tech: 開源模型技術 / Open LLM Tech: Core Technologies
---

本軌聚焦於大型語言模型（LLM）的底層核心技術與開發流程，而非最終生成的內容。我們歡迎關於模型訓練、架構優化與部署實務的技術分享。主題包括但不限於：

1. 模型架構與訓練實務
探討模型結構設計、預訓練（Pre-training）、微調（Fine-tuning）技術如 LoRA 或 QLoRA，以及如何提升訓練效率。
2. 模型量化與布署優化
分享模型量化技術、推論引擎（如 llama.cpp, vLLM）的優化，以及如何在不同硬體環境中達成高效能部署。
3. 資料集處理與開放標準
討論高品質訓練資料的清洗、標記流程，以及確保資料集與模型權重符合開放授權規範的實務。

This track focuses on the underlying core technologies and development workflows of Large Language Models (LLMs), rather than the generated output. We welcome technical submissions regarding model training, architectural optimization, and deployment practices. Topics include, but are not limited to:

1. Model Architecture & Training Practices
Explore model structural design, pre-training, and fine-tuning techniques such as LoRA or QLoRA, and methods to improve training efficiency.
2. Quantization & Deployment Optimization
Share insights on model quantization, optimization of inference engines (such as llama.cpp, vLLM), and achieving high-performance deployment across various hardware environments.
3. Dataset Processing & Open Standards
Discuss the cleaning and labeling processes for high-quality training data, and practices to ensure datasets and model weights comply with open-licensing standards.