Harry Dong

Generalized Parallel Scaling with Interdependent Generations

STEM: SCALING TRANSFORMERS WITH EMBEDDING MODULES

Scalable LLM Math Reasoning Acceleration with Low-rank Distillation

ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference

ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference