跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 26 天前AI 評分53

SO-101 Bench:面向低成本機械臂的語言條件空間操作基準

SO-101 Bench: Accessible Spatial Manipulation

AI 導讀

SO-101 Bench 是以真實 SO-101 機械臂數碼孿生建立的模擬基準,用於評估語言條件下的空間操作。基準涵蓋 138 件物件及 916 個評估回合;五種未使用真實數據訓練的策略在模擬與實機測試中的排名完全一致(Spearman rho = 1.00,MMRV = 0.000),但模擬成功率偏高。

正文

Published on Oct 27

Authors:

Abstract

A simulated benchmark for language-conditioned spatial manipulation on low-cost robot arms reveals that hybrid LLM agents outperform learned vision-language-action policies, with ranking consistency between simulation and real hardware.

Low-cost arms such as the SO-101 have become the most common entry point to robot learning, yet no released policy performs spatially constrained, language-conditioned tasks on them out of the box, and existing SO-101 benchmarks run only on physical hardware with simple tasks and few held-out objects. We introduce SO-101 Bench, a simulated benchmark for language-conditioned spatial manipulation built on an Isaac Lab digital twin of a real SO-101 setup. It has four tasks scored as five families, 138 objects in seen, unseen-instance and unseen-class splits, 916 evaluation episodes with subsets for held-out appearances, rotate-to-fit planning and occlusion memory, and 3,629 simulated demonstrations on which every learned policy is fine-tuned under an equal compute budget. The twin ranks policies as the real robot does: five policies trained without real data, four VLAs and an LLM agent, keep their exact ranking when the same weights run on the physical arm in an unseen environment with unseen objects (Spearman rho = 1.00, MMRV = 0.000), although simulation overestimates success. Across nine VLAs and world-action models, the best learned policies score near 61%, with spatial placement, unseen object classes and rotate-to-fit planning remaining open. An LLM agent that writes code and hands grasping to a learned policy scores 83.2%, and an ablation shows that this lead comes from the choice of harness and the choice of language model.

View arXiv page View PDF Project page GitHub Add to collection

Get this paper in your agent:

hf papers read 2610.36420

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/2610.36420 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/2610.36420 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/2610.36420 in a Space README.md to link it from this page.

Collections including this paper 1

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