Aditya Akella

CONGO: Compressive Online Gradient Optimization

HALoS: Hierarchical Asynchronous Local SGD over Slow Networks for Geo-Distributed Large Language Model Training

On-the-Fly Adaptive Distillation of Transformer to Dual-State Linear Attention for Long-Context LLM Serving

Large Language Models as Realistic Microservice Trace Generators

$Read-ME$: Refactorizing LLMs as Router-Decoupled Mixture of Experts with System Co-Design