@misc{lin2023activation, title={AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration}, author={Ji Lin and Jiaming Tang and Haotian Tang and Shang Yang and Xingyu Dang and Song Han}, year={2023}, eprint={2306.00978}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2306.00978} }