G-CORE:一个简单、可扩展且均衡的 RLHF 训练器.pdfVIP

  • 5
  • 0
  • 约7.23万字
  • 约 20页
  • 2026-03-09 发布于北京
  • 举报

G-CORE:一个简单、可扩展且均衡的 RLHF 训练器.pdf

G-CORE:一个简单、可扩展且均衡的RLHF训练器

JunyuWuWeimingChangXiaotaoLiu

TencentTencentTencent

nrwu@astrachang@xiaotaoliu@

GuanyouHeHaoqiangHongBoqiLiu

TencentTencentTencent

guanyouhe@jeffhong@parkeychen@

本HongtaoTianTaoYangYunshengShi

译TencentTencent

文档评论(0)

1亿VIP精品文档

相关文档