collaborators

5 papers

cs.LG2026

Staleness-Learning Rate Scaling Laws for Asynchronous RLHF

Jingwei Song, Haofeng Xu, Jie Xiao +8

High-throughput RLHF systems often decouple rollout generation from policy optimization, leading to the use of stale rollouts during learner updates. In this work, we study the eff…

cs.LG2026

ECHO-2: A Large-Scale Distributed Rollout Framework for Cost-Efficient Reinforcement Learning

Jingwei Song, Meng Chen, Jie Xiao +15

Reinforcement learning (RL) is a critical stage in post-training large language models (LLMs), involving repeated interaction between rollout generation, reward evaluation, and cen…

cs.LG2026

MARS: Unleashing the Power of Speculative Decoding via Margin-Aware Verification

Jingwei Song, Xinyu Wang, Hanbin Wang +6

Speculative Decoding (SD) accelerates autoregressive large language model (LLM) inference by decoupling generation and verification. While recent methods improve draft quality by t…

cs.LG2026

Co-Evolution of Policy and Internal Reward for Language Agents

Xinyu Wang, Hanwei Wu, Jingwei Song +8

Large language model (LLM) agents learn by interacting with environments, but long-horizon training remains fundamentally bottlenecked by sparse and delayed rewards. Existing metho…

cs.DC2025

Speculative Decoding in Decentralized LLM Inference: Turning Communication Latency into Computation Throughput

Jingwei Song, Wanyi Chen, Xinyuan Song +7

Speculative decoding accelerates large language model (LLM) inference by using a lightweight draft model to propose tokens that are later verified by a stronger target model. While…