paper-with-me

홈 › Papers

Eq.Bot: Enhance Robotic Manipulation Learning via Group Equivariant Canonicalization

2025-11-19 · Jian Deng, Yuandong Wang, Yangfu Zhu, Tao Feng, Tianyu Wo, Zhenzhou Shao arxiv

Robotic manipulation systems are increasingly deployed across diverse domains. Yet existing multi-modal learning frameworks lack inherent guarantees of geometric consistency, struggling to handle spatial transformations such as rotations and translations. While recent works attempt to introduce equivariance through bespoke architectural modifications, these methods suffer from high implementation complexity, computational cost, and poor portability. Inspired by human cognitive processes in spatial reasoning, we propose Eq.Bot, a universal canonicalization framework grounded in SE(2) group equivariant theory for robotic manipulation learning. Our framework transforms observations into a canonical space, applies an existing policy, and maps the resulting actions back to the original space. As a model-agnostic solution, Eq.Bot aims to endow models with spatial equivariance without requiring architectural modifications. Extensive experiments demonstrate the superiority of Eq.Bot under both CNN-based (e.g., CLIPort) and Transformer-based (e.g., OpenVLA-OFT) architectures over existing methods on various robotic manipulation tasks, where the most significant improvement can reach 50.0%.

📄 PDF Abstract BibTeX arXiv:2511.15194

Code (0)

등록된 구현이 없습니다.

Tasks

Spatial Reasoning

Similar Papers 제목 키워드 기반

Lie Algebra Canonicalization: Equivariant Neural Operators under arbitrary Lie Groups

2024-10-03 · Zakhar Shumaylov, Peter Zaika, James Rowbottom, Ferdia Sherry 외

The quest for robust and generalizable machine learning models has driven recent interest in exploiting symmetries through equivariant neural networks. In the context of PDE solvers, recent works have shown that Lie poin…

image-classificationImage ClassificationInductive Bias

Equivariance with Learned Canonicalization Functions

2022-11-11 · Sékou-Oumar Kaba, Arnab Kumar Mondal, Yan Zhang, Yoshua Bengio 외

Symmetry-based neural networks often constrain the architecture in order to achieve invariance or equivariance to a group of transformations. In this paper, we propose an alternative that avoids this architectural constr…

image-classificationImage ClassificationPoint Cloud Classification

A Canonicalization Perspective on Invariant and Equivariant Learning

2024-05-28 · George Ma, Yifei Wang, Derek Lim, Stefanie Jegelka 외

In many applications, we desire neural networks to exhibit invariance or equivariance to certain groups due to symmetries inherent in the data. Recently, frame-averaging methods emerged to be a unified framework for atta…

Graph ClassificationGraph EmbeddingGraph Regression

Equivariant Descriptor Fields: SE(3)-Equivariant Energy-Based Models for End-to-End Visual Robotic Manipulation Learning

2022-06-16 · Hyunwoo Ryu, Hong-in Lee, Jeong-Hoon Lee, Jongeun Choi

End-to-end learning for visual robotic manipulation is known to suffer from sample inefficiency, requiring large numbers of demonstrations. The spatial roto-translation equivariance, or the SE(3)-equivariance can be expl…

Lorentz Local Canonicalization: How to Make Any Network Lorentz-Equivariant

2025-05-26 · Jonas Spinner, Luigi Favaro, Peter Lippmann, Sebastian Pitz 외

Lorentz-equivariant neural networks are becoming the leading architectures for high-energy physics. Current implementations rely on specialized layers, limiting architectural choices. We introduce Lorentz Local Canonical…

Data Augmentation