跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-20AI 評分43

S-Agent 2.0:通用編碼智能體的空間推理

S-Agent 2.0: Spatial Reasoning with a General Coding Agent

AI 導讀

S-Agent 2.0 為通用編碼智能體整合持久場景記憶、自然語言解讀及 3D 感知工具,以提升空間推理能力。在 MMSI-Bench 和 ReVSI 上,搭配 Qwen3.8-27B 分別取得 66.5 和 82.3 分,搭配 Gemini-3.8-Flash 則取得 74.1 和 90.5 分。它亦可配合 Codex 和 Claude Code 等不同通用編碼智能體框架使用。

正文

Published on Oct 20

·

Submitted by

leoli

on Oct 21

Authors:

,

,

,

,

,

,

,

,

,

,

Abstract

S-Agent 2.0 improves spatial reasoning by integrating a persistent scene memory and plain-language interpretation into a coding agent with 3D perception tools.

Spatial reasoning asks a model to work out where objects are in 3D space, how they relate to one another, and how they move. Agentic VLMs have recently brought large improvements on such tasks by letting the model actively gather evidence with 3D perception tools instead of answering in a single pass. Even so, on many questions the tools return the right evidence and the model still does not arrive at the correct answer: the gap between model and tools, where raw readings become usable conclusions, is left unfilled. We present S-Agent 2.0, a spatial reasoning agent that equips an off-the-shelf general coding agent with a command-line perception toolkit, and we design the interface around one principle: help the VLM close the evidence loop: acquire scene evidence, remember it in a persistent scene memory, and interpret each reading in plain spatial language. This design keeps the full flexibility of a coding agent, and every reading it acts on stays grounded in the tools' geometry. On MMSI-Bench and ReVSI, the toolkit improves a strong open-source model, Qwen3.8-27B, by +13.9 and +20.5 points over the same agent without tools, and a strong closed-source model, Gemini-3.8-Flash, by +11.2 and +9.6 points. With the open backbone S-Agent 2.0 reaches 66.5 on MMSI-Bench and 82.3 on ReVSI, ahead of the specialized system SpatialClaw on the same backbone (62.6 and 74.5), and with Gemini-3.8-Flash it reaches 74.1 and 90.5 against its 67.0 and 79.7. Both tool gains shrink as the backbone grows more capable, but SpatialClaw's shrinks far more on ReVSI, from +13.2 to +2.7, while ours stays at +9.6. S-Agent 2.0 also works across different general coding agent harnesses, such as Codex and Claude Code.

View arXiv page View PDF Project page GitHub Add to collection

Get this paper in your agent:

hf papers read 2610.28354

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.28354 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.28354 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.28354 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co