# Fudan and NTU propose SimpleVLN, using a native multimodal large model to set a new SOTA in vision-language navigation

- Source: 机器之心 · 微信公众号
- Published: 2026-10-30T01:31:15.000Z
- RoboHot: https://robohot.news/en/items/opphgypj9gr4hg3btayy4jkhq
- Original: https://mp.weixin.qq.com/s?__biz=MzA3MzI4MjgzMw==&mid=2651071038&idx=3&sn=a8d84743153b0e1e61fc73fd5cc7e261&chksm=85b5c200521bbd40f555b838005377226ccbd86ccbf8e988bf5dd85649628a8b394243693a8d&scene=126&sessionid=1793371799#rd

## Summary

A team from Fudan University and Nanyang Technological University, Singapore, proposed SimpleVLN, which performs vision-language navigation without changing the core structure of the multimodal large model or adding dedicated navigation modules, relying only on historical frame selection, action-chunk decision-making, and three-stage training with SFT, DAgger, and reinforcement learning.