# Preprint improves robot navigation model while removing runtime planner

_Published Friday, November 13, 2026 at 1:30 PM EST · Robotics, AI, Science · Latest · Tier 2 — Notable_

Researchers report that a training method raised an 8B navigation model's success rate from 35.0% to 44.9% across all 3,189 REVERIE-CE val-unseen episodes, while allowing it to navigate without a large language model planner at test time.

The method, H2E, turns successful agent sessions into training examples for a model that converts visual and language inputs into actions. It excludes training penalties for steps moving away from the goal and labels visited states within 1 m of the goal as stopping points. The findings are presented in a preprint and concern the specified navigation benchmark.

## Sources

- [arXiv Query: search_query=cat:cs.RO&id_list=&start=0&max_results=30](https://arxiv.org/abs/2611.16246v1)

---
Canonical: https://techandbusiness.org/newswire/_beXiuLleaO-hZrZXjIW9R
Published: 2026-11-13T18:30:39.960Z
Story chronology: 2026-11-12T03:42:58.000Z
Retrieved: 2026-11-13T21:20:48.782Z
Publisher: Tech & Business (techandbusiness.org)