---
title: "SCOUT：通过结构化思维链与多目标过程奖励增强空间推理"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.12220v1"
permalink: "https://feed.astralor.com/items/itm_944207d271cea00ea0c4cc13"
published: "2026-08-12T16:14:25.000Z"
score: 76
channel: "ai"
tags: ["SCOUT","强化学习","思维链","空间推理","视觉语言模型"]
---

# SCOUT：通过结构化思维链与多目标过程奖励增强空间推理

## AI 点评

> 空间推理是VLM从“看见”走向“理解”的关键一跳。SCOUT把结构化思维链和多目标奖励结合，对做具身智能、自动驾驶或视觉导航的团队尤其有参考价值；建议关注其数据集与训练细节，看看能否迁移到自己的场景。

## 速览

现有视觉语言模型（VLM）在空间推理上存在明显瓶颈。近期强化学习方法虽以可验证结果弥补差距，但中间步骤信用分配不佳，且结构化推理忽视3D深度感知。为此，研究者提出SCOUT，设计显式建模3D环境感知的结构化思维链框架，并引入多目标过程奖励与定制优势估计的强化学习算法，实现细粒度信用分配并构建SCOUT-24k数据集。实验表明，SCOUT-3B在通用空间基准和复杂空间推理任务上分别提升16.85%和6.3%，SCOUT-7B超越GPT-4o达4.28%，并展现出多图像和视频场景的泛化能力。

## 关键点

- 提出SCOUT框架，采用结构化思维链显式建模3D环境感知。
- 引入多目标过程奖励RL算法，实现推理轨迹的细粒度信用分配。
- 构建SCOUT-24k结构化空间推理CoT数据集。
- 实验显示SCOUT-7B在空间基准上超越GPT-4o 4.28%，并泛化至多图像和视频场景。

## 全文

Existing Vision-Language Models (VLMs) exhibits a critical bottleneck in robust spatial reasoning. Recent reinforcement learning (RL) methods aim to close this gap with verifiable outcomes, yet they suffer from poor credit assignment across intermediate reasoning steps. Concurrently, structured reasoning approaches overlook the critical depth perception necessary for comprehensive 3D understanding. To address these challenges, we propose SCOUT (Structured Chain-Of-Thought Utilizing Process-Supervised RL Training). Specifically, we design a structured Chain-of-Thought (CoT) framework that explicitly models 3D environmental perception to ensure robust spatial understanding and reasoning. Furthermore, we introduce a novel RL algorithm featuring multi-objective process rewards and a tailored advantage estimation method, facilitating fine-grained credit assignment across distinct segments of the reasoning trajectory. To support our framework, we develop SCOUT-24k, a structured spatial reasoning CoT dataset synthesized through a customized pipeline. Extensive evaluations demonstrate that SCOUT-3B improves upon baseline models by 16.85% and 6.3% on general spatial benchmarks and complex spatial reasoning tasks respectively. Notably, our larger SCOUT-7B even outperforms GPT-4o by a margin of 4.28%. Moreover, despite being trained exclusively on single image, SCOUT-7B exhibits robust out-of-domain generalization to multi-image and video scenarios. These empirical results render SCOUT as a critical step towards next generation of spatially-aware VLMs.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_944207d271cea00ea0c4cc13