TRACE统一FP4训练与Rollout量化
AI 总结
阿里千问与俄亥俄州立大学研究人员提出TRACE，利用Rollout侧真实的FP4量化结果指导训练侧舍入，以减少MoE语言模型强化学习中的训练—生成路径差异。论文在四个模型的推理、编程和长程RL任务上评估后，作者报告联合FP4权重、激活与KV Cache的Rollout可取得接近BF16的表现，并实现最高5.4倍Rollout加速。框架还通过缓存部分尾数与缩放信息，降低传输Rollout指导数据的开销。
深度见解
论文显示，对齐训练与Rollout的量化路径可在保持接近BF16的RL性能的同时加速FP4生成，直接针对MoE模型后训练中的高成本环节。
Computer Science > Machine Learning
arXiv:2610.07767 (cs)
[Submitted on 6 Oct 2026]
Title:TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models
Authors:Xin Wang, Hao Yu, Zhengyang Zhuge, Bochao Mao, Zheng Li, Junda Feng, Yuyan Luo, Yi Zhang, Yizhong Cao, Mi Zhang, Dayiheng Liu, Jianwei Zhang
View a PDF of the paper titled TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models, by Xin Wang and 11 other authors
#TRACE#Alibaba Qwen#Ohio State University#Qwen3.5-35B-A3B#Qwen3.5-122B-A10B#Qwen3.8-Flash-Next#Qwen3.8-2.4T-A95B