# Yichuan Wang团队在TorchTitan RL和vLLM上为Gated DeltaNet实现了训练与推理的逐位精确匹配，并提出了异步RL中该匹配是否有效的问题

- 来源：Nathan Lambert
- 发布时间：2026-08-07 21:54
- AIWatch 分数：64
- AIWatch 标记：未精选
- AIWatch 链接：https://aiwatch.icu/events/evt_01kze969t2m1ed8x6ptbhs0rnw
- 原文链接：https://x.com/natolambert/status/2085726242314346760

## 精选理由

常规快讯，保留列表

## AI 摘要

Yichuan Wang团队在TorchTitan RL和vLLM上为Gated DeltaNet实现了训练与推理的逐位精确匹配，并提出了异步RL中该匹配是否有效的问题。

## 正文

Yichuan Wang: Zero Train–Inference Mismatch — now for linear attention, and under async RL 🎯

We got bitwise-exact trainer/generator parity for Gated DeltaNet (Qwen3.5-9B / 35B-A3B) on TorchTitan RL + vLLM, then asked the question nobody had actually tested in open source: does it help async