TY  - RPRT
TI  - Optimizing LLM Inference Throughput via Memory-aware and SLA-constrained Dynamic Batching
AU  - Bowen Pang
AU  - Kai Li
AU  - Feifan Wang
PY  - 2025
UR  - https://arxiv.org/abs/2503.05248
ID  - 2503.05248
ER  -