@article{sledaspeculativellmdecodingframeworkfor, title = {SLED: A Speculative LLM Decoding Framework for Efficient Edge Serving}, author = {Xiangchen Li and Dimitrios Spatharakis and Saeid Ghafouri and Jiakun Fan and Hans Vandierendonck and Deepu John and Bo Ji and Dimitrios Nikolopoulos}, year = {2025}, eprint = {2506.09397}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2506.09397v2}, }