@article{hybridjitcudagraphoptimizationforlowlate,
  title = {Hybrid JIT-CUDA Graph Optimization for Low-Latency Large Language Model Inference},
  author = {Divakar Kumar Yadav and Tian Zhao},
  year = {2026},
  eprint = {2604.23467},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2604.23467},
}