IsoAct is a post‑hoc debiasing framework that respects the geometry of non‑Euclidean representations. It identifies nuisance‑predictive directions via Fisher discriminant analysis in tangent coordinates and corrects them using manifold‑native actions that keep the representation on its manifold. Experiments on hyperspherical, hyperbolic, and SE(3) spaces show that IsoAct reduces nuisance recoverability while preserving utility and manifold validity, notably improving kinase‑group recoverability and structural prediction on the KLIFS dataset.
By Sumin Park, Taero Kim, Minhyeong Cho, Subeen Park, Jong Hwan Lee, Kyungwoo Song
Semantic-preserving transformations can induce substantial motion in learned representations, while small changes may strongly affect model predictions, raising a basic question: what local metric bes...
arXiv:2609.36458v1 Announce Type: new
Abstract: Semantic-preserving transformations can induce substantial motion in learned representations, while small changes may strongly affect model predictions...
By Abdullah All Tanvir, Xin Zhong
arXiv:2602.02611v2 Announce Type: replace
Abstract: A prevailing paradigm in modern representation learning is the map-first approach, in which a representation map is learned from reconstruction, em...
By David Vigouroux (ANITI, IMT Atlantique - DSD, LaTIM), Lucas Drumetz (IMT Atlantique - MEE, Lab-STICC\_OSE, ODYSSEY), Ronan Fablet (IMT Atlantique - MEE, Lab-STICC\_OSE, ODYSSEY), Fran\c{c}ois Rousseau (IMT Atlantique - DSD, LaTIM)
The paper introduces a Nested Inductive Bias framework that uses a two‑stage diffeomorphic composition to pull back non‑Euclidean target geometries onto symmetric positive definite (SPD) manifolds. This approach allows the construction of curvature‑aligned Riemannian classifiers that respect both matrix constraints and the intrinsic relational geometry of data. Empirical results on kinematic, signal processing, and synthetic benchmarks show that class separability degrades when metric curvature does not match the data distribution, and the authors also propose the Rational Conformal Metric (RCM) for robust vectorized architectures.
By Tushar Das
arXiv:2610.29168v1 Announce Type: cross
Abstract: World action models (WAMs) are shown to substantially diverge under generalization performance based on their generative backbones. We investigate th...
By Junha Chun, Youngjoon Jeong, Taesup Kim
arXiv:2207. 03116v4 Announce Type: replace Abstract: We introduce a general method for learning representations that are equivariant to symmetries of data.
By Giovanni Luca Marchetti, Gustaf Tegn\'er, Anastasiia Varava, Danica Kragic
arXiv:2609.15083v1 Announce Type: new
Abstract: Mixed-curvature representation learning seeks to capture rich geometric structures that cannot be adequately modeled by a single curvature regime. Exis...
By Xingrun Li, Yusuke Mukuta, Xin Yang, Yinyu Ye, Tatsuya Harada
arXiv:2505. 19809v3 Announce Type: replace-cross Abstract: In many real-world applications of regression, conditional probability estimation, and uncertainty quantification, exploiting symmetries rooted in physics or geometry can dramatically improve generalization and sample efficiency.
By Daniel Ordo\~nez-Apraez, Vladimir Kosti\'c, Alek Fr\"ohlich, Vivien Brandt, Karim Lounici, Massimiliano Pontil
arXiv:2610.19007v1 Announce Type: new
Abstract: World models are a powerful means of reasoning and planning, but many of these action-conditioned predictive models are high dimensional and computatio...
By Anton Arapin, Chahyon Ku, Bernadette Bucher
arXiv:2607. 08987v1 Announce Type: new Abstract: Spectral embedding methods are widely used for dimensionality reduction and clustering of high-dimensional datasets with intrinsic low-dimensional structures.
By Yeari Vigder, Paulina Hoyos, David Thong, Joakim and\'en, Joe Kileel, Amit Moscovich
The paper extends the analysis of Joint-Embedding Predictive Architectures (JEPAs) beyond Euclidean latent spaces to Riemannian manifolds. It shows that when latent variables lie on a sphere and the target distribution matches this spherical geometry, every optimal representation recovers the latent state up to an orthogonal transformation, demonstrating that Gaussian uniqueness is not universal. Experiments confirm that geometrically compatible targets improve linear recovery, especially in high-dimensional toroidal settings.
By L\'eo Nicollier (CB, ATT), Enric Meinhardt-Llopis (CB), Marc Pic (ATT), Pablo Mus\'e (CB, IFUMI), Gabriele Facciolo (CB)