Equivariant Latent Alignment via Flow Matching under Group Symmetries
Quick Take
The paper introduces Residual Latent Flow, a flow-based framework that addresses latent misalignment in equivariant representation learning, significantly enhancing novel view synthesis quality under rotation groups SO(n). Experiments demonstrate improved compliance with group symmetries, leading to better visual fidelity and consistency.
Key Points
- Residual Latent Flow corrects latent misalignment in equivariant representation learning.
- The method improves novel view synthesis quality significantly under SO(n) rotation groups.
- Experiments show enhanced compliance with underlying group symmetries.
- Geometry-aware generative models benefit from improved interpretability and generalization.
Article Excerpt
From source RSS / original summaryarXiv:2605. 30705v1 Announce Type: new Abstract: Geometry-aware generative models and novel view synthesis approaches have shown strong potential in visual fidelity and consistency. In parallel, equivariant representation learning has emerged as a powerful framework for constructing latent spaces where analytically known group transformations could act directly, capturing geometric structure in data and enhancing both interpretability and generalization in novel view synthesis.
However, we identify that existing approaches often suffer from latent misalignment, a discrepancy between the intended group action and the actually required transformations in the latent space. Consequently, the learned latents often fail to consistently preserve the equivariant relations imposed by the underlying group symmetry. To address this, we propose Residual Latent Flow, a flow-based framework that corrects the misaligned latents, thereby improving compliance with the underlying equivariance relation.
Our comprehensive experiments show that our method significantly reduces latent misalignment and improves novel view synthesis quality, under rotation groups SO(n).
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →Evi-Steer: Learning to Steer Biomedical Vision-Language Models through Efficient and Generalizable Evidential Tuning
Evi-Steer introduces a novel evidential tuning framework for BiomedCLIP, enabling efficient fine-tuning with only 0.11% parameter updates. It significantly enhances performance in few-shot learning and domain shifts across 15 biomedical imaging datasets, demonstrating robustness for clinical applications.