Xiao Chen's Learning Space

Welcome to Xiao Chen's Learning Space! This is a personal technical hub dedicated to the exploration and study of cutting-edge technologies in Large Models.

Here, we are committed to deeply understanding and practicing core technologies related to large models, covering the complete pipeline from foundational training to high-level applications. The main areas of study and research include:

• Pre-training: Exploring the foundational construction of large models, researching massive data processing and base model training mechanisms.
• Fine-tuning: Studying techniques like SFT and LoRA to better adapt large models to vertical domains and specific tasks.
• Reinforcement Learning: Exploring alignment techniques such as RLHF and DPO to improve the safety and human-preference consistency of model outputs.
• Agent: Researching autonomous agents based on large models, exploring planning, memory, tool usage, and multi-agent collaboration.
• Vision-Language Models (VLM): Delving into multimodal technologies, researching the fusion and understanding of visual and linguistic features.
• Vision-Language-Action Models (VLA): Focusing on the frontier of embodied AI, researching the joint modeling and execution of vision, language, and physical actions.

Xiao Chen's Learning Space aims to document the learning journey and share technical insights. We look forward to progressing together with like-minded partners and witnessing the rapid development of AI technology!