tracked author

Chi Zhang

HybridFlow, DAPO, and Laminar recurring author; GitHub and homepage align with USC-to-ByteDance RLHF infrastructure work, but no high-confidence personal X account was found.

3 archived notes X: not-found Homepage

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。

归档 Jun 28, 2026

Laminar: A Scalable Asynchronous RL Post Training Framework

让完成轨迹独立进入 experience buffer，并以 CPU/RDMA relay 和同版本 repack 解除全局 batch / 权重同步 barrier。

待审阅 2510.12633-laminar-asynchronous-rl-post-training RL Infrastructure Rollout Optimization Distributed Training

归档 Mar 03, 2026

DAPO: An Open Source LLM Reinforcement Learning System at Scale

用 Clip Higher、Dynamic Sampling、token level loss 与 overlong shaping 构成可复现 long CoT RL recipe。

待审阅 2503.14476-dapo-long-cot-rl-system RL Algorithm Reasoning RL RL Infrastructure

归档 Jan 27, 2026

HybridFlow: A Flexible and Efficient RLHF Framework

用跨模型 single controller 与模型内 multi controller 统一编排 RLHF dataflow 和并行执行。

待审阅 2409.19256-hybridflow-rlhf-framework RL Infrastructure Distributed Training Rollout Optimization