Dmytro Mishkin ๐Ÿ‡บ๐Ÿ‡ฆ (@ducha_aiki)

2025-01-21 | โค๏ธ 66 | ๐Ÿ” 15


SpatialCoT: Advancing Spatial Reasoning through Coordinate Alignment and Chain-of-Thought for Embodied Task Planning

Yuecheng Liu et 13 al tl;dr: train a LoRA for VLM to make it understand in-image coordinates first, then plan for the navigation https://arxiv.org/abs/2501.10074 https://x.com/ducha_aiki/status/1881658788316635341/photo/1

๐Ÿ”— ์›๋ณธ ๋งํฌ

๋ฏธ๋””์–ด

photo

photo

photo

photo


Auto-generated - needs manual review

Tags

domain-robotics domain-vlm domain-visionos