SyncAI.news, a Varaisys broadcasting
CascadeJEPA: Enhancing 3D Geometric Representations via Spatially Coupled Cascaded Prediction
KZ

Kailai Zhou, Ziqi Chen, Xiaoning Xu, Yuefeng Wu, Zixuan Fu, Zijun Lin, Zhihao Li, Youwei Pang, Xiaoqi Zhao, Xun Cao, Bihan Wen

· 1 min read

ResearcharXiv cs.CV

CascadeJEPA: Enhancing 3D Geometric Representations via Spatially Coupled Cascaded Prediction

arXiv:2611.22686v1 Announce Type: new Abstract: Joint-Embedding Predictive Architectures (JEPAs) have emerged as a promising paradigm for predictive visual representation learning. However, our analysis of I-JEPA reveals a growing discrepancy between semantic and geometric learning: improved semantic recognition during pretraining is accompanied by degraded 3D geometric representations. We hypothesize that this partly stems from predicting highly abstract latent representations with incomplete spatial context, favoring semantic regularities over fine-grained geometric information. To address this issue, we introduce CascadeJEPA, a spatially coupled cascaded prediction framework that promotes richer patch-level interactions through full-grid context encoding and cross-attention between prediction stages, while largely preserving the original I-JEPA architecture. Extensive experiments demonstrate substantial improvements over I-JEPA across multiple 3D geometric tasks while maintaining strong semantic recognition performance. Furthermore, its advantages extend to future feature forecasting and world model planning, highlighting the potential of enhanced geometric representations to support predictive world modeling.

Original source

This story was published by arXiv cs.CV and written by Kailai Zhou, Ziqi Chen, Xiaoning Xu, Yuefeng Wu, Zixuan Fu, Zijun Lin, Zhihao Li, Youwei Pang, Xiaoqi Zhao, Xun Cao, Bihan Wen. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News