Jinri Yao contributed to the vllm-project/llm-compressor repository by addressing a documentation inconsistency in the Qwen3-VL AWQ example. Focusing on data processing and machine learning workflows, Jinri identified that the save directory name did not accurately reflect the quantization configuration, which could hinder reproducibility and experiment tracking. Using Python scripting, Jinri updated the example script, comments, and directory naming to align with the actual 4-bit weight (W4A16) quantization used in the AWQ recipe. This work maintained functional stability while improving clarity and consistency, demonstrating careful attention to detail and a commitment to reliable, reproducible machine learning practices.