tracked author

Junxiong Wang

Coauthor of ThunderAgent. Homepage identifies him as leading several Together AI projects including adaptive speculative decoding, inference-time training, and efficient RL rollouts, with a Cornell CS PhD background.

1 archived notes X: high HomepageX

Related Notes

按论文归档时间排序,展示该作者在本站已经出现的材料。

Jul 03, 2026

ThunderAgent: A Simple, Fast and Program Aware Agentic Inference System

ThunderAgent 把跨多轮 LLM 调用与工具等待建模为带 phase、status、KV footprint、backend placement 和 tool dependency 的 LLM Program,使 scheduler 能按 Reasoning / Acting 状态控制 KV working set,并在释放 KV 后通过全局队列重新选择恢复节点;论文在所测高并发 tool use serving 与 rol...

2602.13692-thunderagent-program-aware-agentic-inference Agent WorkflowInference SchedulingKV Cache