km1994/llms_paper

Comprehensive study notes and analysis repository systematizing top-tier conference papers on LLM engineering, organized across domain-specific sections including multimodal vision-language models (Gemini, LLaVA, GPT-4V), parameter-efficient fine-tuning techniques, retrieval-augmented generation applications across medical/legal/knowledge domains, and reasoning approaches like chain-of-thought. The collection features detailed paper summaries with methodology explanations, experimental results, and GitHub implementations, enabling practitioners to quickly understand key innovations in model architectures, training strategies, and task-specific adaptations. Spans research from foundation model pretraining and inference optimization to agent design, knowledge graph integration, and evaluation methodologies across Chinese and English publications.

373 stars. No commits in the last 6 months.