Neural Representation Geometry

Follow this topic Ask about this topic

Momentum

18 papers in the last four weeks, up 125% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 182

Aug 7, 2026cs.CL

Calibrating WEAT Against Anisotropy: ZCA Whitening as a Geometric Pre-Processing Step for Embedding Association Tests

We propose Zero-phase Component Analysis (ZCA) whitening as a geometric pre-processing step for the Word Embedding Association Test (WEAT). WEAT is a bias measurement method widely used in both computational social science and AI fairness research. It relies on cosine similarity as a measure of semantic association, which assumes that the embedding space is approximately isotropic. However, prior work has reported that many widely used language models do not satisfy this assumption, raising concerns about the reliability of bias measurements. ZCA whitening transforms the covariance of the embedding space into the identity matrix while minimizing perturbation to the original vectors. This transformation restores the isotropy condition on which WEAT relies. We evaluate our approach on ten standard WEAT test suites and seven models spanning three architectural families, yielding 70 model-task combinations. The results show that ZCA whitening substantially reduces the anisotropy of the embedding spaces across all models. Particularly for highly anisotropic models, we further observe improvements on standard semantic similarity benchmarks, indicating that the calibrated space better captures semantic associations. After calibration, over 30% of WEAT results change significance status, and effect sizes shift in both directions depending on bias category. These shifts suggest that uncalibrated measurements may both overestimate and underestimate the associations encoded in the embedding space. These findings indicate that previously reported bias measurements in anisotropic embedding spaces should be interpreted with caution and may benefit from re-evaluation with calibrated methods. Our approach contributes to restoring the measurement foundation of WEAT across both computational social science and AI fairness research.

Seitaro Ono, Senna Ross, Jun Saiki

Graduate School of Human and Environmental Studies, Kyoto University, Kyoto, Japan · Brock University, St. Catharines, Ontario, Canada

Neural Representation Geometry Text Embedding Evaluation Language Model Bias Evaluation

Aug 4, 2026cs.SD

On the Geometry of Music Bandwidth Extension in Latent Spaces of Audio Codecs

Recent audio restoration increasingly relies on large-scale conditional latent generative modeling, including diffusion, Schrodinger Bridges, and Flow Matching variants, to invert degradations such as bandwidth limitation or noise. We present an analysis of the performance of various state-of-the-art methods compared to simple arithmetic transformations in the latent spaces of multiple neural codecs for musical bandwidth extension. We show that estimating a single transport vector between the clean and degraded latent centroids on a reference set, and adding it to degraded latents, can yield restoration performance competitive with large diffusion models. This suggests, first, that some neural codec latent spaces exhibit structure aligned with audio bandwidth; and second, that in such cases complex conditional models may offer only limited gains over a simple vector addition. We argue that these findings reveal an interesting avenue for future research whereby models could take advantage of the latent space structure in order to offer greater training and parameter efficiency, and overall better performance. Additionally, we propose to consider this simple arithmetic transformation as a baseline for music bandwidth extension research, as it allows an assessment of the contribution of learnable parameters towards restoration performance.

Hendrik Vincent Koops, Hao Hao Tan, Elio Quinton

Music & Audio Machine Learning Lab · Universal Music Group, London, U.K.

Neural Audio Codecs Neural Representation Geometry Audio Diffusion Models

Aug 3, 2026cs.CV

Transformer Geometry Observatory TGO-III: Semantic Geometry Observatory

With the widespread adoption of Vision Transformers in modern AI, the need to analyze their inherent representational behavior has become increasingly important. While most existing studies emphasize token geometries and training dynamics, the evolution of representational covariance structures and class-level geometric organization remains comparatively underexplored. In this work, we investigate semantic geometry and class separability as representations evolve across the layers of ViT-Small/16 through TGO-III: Semantic Geometry Observatory. It is a framework designed to analyze the emergence of semantic organization, feature evolution, and class-wise representation geometry throughout training. The framework employs multiple complementary observatories, including Linear Probe Accuracy, Fisher Ratio, Class Centroid Distances, Local Intrinsic Dimension, and Local PCA Rank, to quantify the progressive evolution of discriminative representations. Our analysis reveals that class representations become progressively more linearly separable, Fisher discriminability increases, class centroids move farther apart, and local representation manifolds exhibit structured class-dependent geometric complexity. These observations provide empirical evidence supporting the Semantic Expansion Hypothesis, suggesting that the manifold expansion observed in previous observatories is accompanied by the progressive organization of representations into increasingly discriminative semantic structures. Collectively, TGO-III extends the Transformer Geometry Observatory framework by establishing a direct connection between manifold geometry, covariance evolution, and semantic organization during Transformer training.

Kaustubh Kapil, Kishor P. Upla

Department of Electronics Engineering Sardar Vallabhai National Institute of Technology (SVNIT), Surat, India

Transformer Interpretability Neural Representation Geometry Vision Transformer

Jul 31, 2026cs.SD

Do Music Foundation Models Embed Pitch in Helical Structure?

This study analyzes the intermediate representations of music foundation models (MFMs) and reports the geometric structures used to represent pitch information. By inputting isolated musical notes into trained MFMs and analyzing their principal components, we reveal that the representations form a helical structure reflecting the octave periodicity of pitch. Furthermore, we show that the clarity and geometry of this helical structure vary not only across models but also with the acoustic properties of the input signals. Our analysis provides a novel approach for clarifying the internal mechanisms of MFMs.

Hayato Yagi, Shinnosuke Takamichi, Rin Sato +2

Keio University, Japan · The University of Tokyo, Japan · Waseda University, Japan

Audio Representation Learning Neural Representation Geometry

Jul 28, 2026stat.ML

More Data, Worse Decisions? Preference Reversals in Neural Networks under Gram Incompatibility

Neural networks increasingly combine data across populations, time periods, and operating conditions to improve generalization. This raises a reliability question: whether a model refitted on pooled data preserves an action ordering supported by both sources. Case-Based Decision Theory (CBDT) formalizes this requirement through its composition axiom, which requires source-supported preferences to survive their union. We study when this property holds for fixed-representation neural networks with ordinary least squares (OLS) output heads. First, we show that pooled refitting recomputes the inverse-Gram geometry used to weight source evidence, which can reverse shared preferences, and derive exact and approximate preservation conditions. Next, we introduce a scale-invariant Gram mismatch measure for prioritizing candidate pools and geometry-oriented regularization for shaping source geometry during training. Finally, we develop a three-stage audit that traces strict pairwise reversals through decision changes to task-defined utility loss. Experiments spanning a load-based bidding proxy and medical and financial decision proxies reveal stable and reversal-prone pooling regimes: the load audit identifies a measurable nonzero class of source-consensus-relative harmful decisions under the proxy utility, while cross-domain audits show that comparable mismatch can correspond to sharply different preservation rates. Geometry-oriented objectives occupy distinct descriptive accuracy-consistency-geometry-harm operating points. Together, the framework makes compositional reliability measurable and operational through screening, analytic certification, geometry-oriented training, and decision-consequence auditing.

Yanli Yan, Yuanzheng Li, Yong Zhao +2

Huazhong University of Science and Technology Wuhan, China

Neural Network Generalization Neural Representation Geometry Decision-Making under Uncertainty

Jul 26, 2026cs.GR

Neural Representation of Minimal Surfaces

We propose a neural representation for minimal surfaces. Unlike prior approaches based on discretization or Physics-Informed Neural Networks (PINNs), where meshes or neural fields are optimized to approximate the governing equations, our method builds on an exact representation, similar to the classical Weierstrass--Enneper parameterization, yielding minimal surfaces up to negligible quadrature error in evaluation. We formulate a training objective for the Plateau problem that optimizes over this representation.

Jiayin Sun, Albert Chern

University of Utah, USA · University of California San Diego, USA

Neural Representation Geometry

Jul 21, 2026cs.LG

Functional Equivalence and Geometric Diversity in Neural Network Approximations: An Empirical Characterization

The Universal Approximation Theorem states that a neural network with a single hidden layer is sufficient to approximate any continuous univariate function on a compact domain to arbitrary error. However, the uniqueness of such neural network representations is not guaranteed, raising questions about practical identifiability. In this work, we address this concern by analyzing functional equivalence and geometric diversity of neural network approximations to a few elementary mathematical functions. The analysis includes an extensive study of single-layer neural networks and multilayer perceptrons under noisy and noise-free conditions. Beyond just network capacity, we study the geometric properties through the lens of sloppiness, characterized by the eigen spectrum of the Hessian of the cost function and the effective rank to quantify the dimensionality of parameter space. The study reveals large equivalence classes of functionally indistinguishable yet geometrically diverse networks that consistently exhibit low effective rank and structural redundancy. Finally, a model select criterion is proposed for identifying optimal models based on parsimony, ease of estimation, and inference efficiency.

Anuragine S A, Prem Jagadeesan

Amrita School of Artificial Intelligence Amrita Vishwa Vidyapeetham, Coimbatore, India

Multilayer Perceptrons Neural Representation Geometry Parameter Identifiability Neural Network Approximation Theory

Jul 20, 2026cs.LG

Topological Signatures of Context-Level Reliability in TabPFN

TabPFN is a transformer-based foundation model for tabular prediction that performs inference without task-specific training by conditioning on a support set and query inputs. Despite its strong empirical performance, its internal behavior on structurally difficult tabular geometries remains poorly understood. We study this behavior using zigzag persistent homology, treating TabPFN layer representations as evolving point clouds. We construct a controlled benchmark of synthetic tabular tasks with known true probabilities and varied intrinsic topology, including warped circles, tori, spheres, Hopf links, trefoil knots, and Swiss rolls. Across these tasks, we find that the topology of TabPFN's internal representation geometry is strongly associated with dataset-level reliability; for example, the zeroth homology group

H_0

fragmentation count correlates positively with mean absolute residual across controlled tasks, and this association strengthens in a high-resolution warped circle case study at large sample size. Harder geometries induce a dual topological signature: increased

H_1

loop activity and increased

H_0

fragmentation, while the

H_1

persistence becomes shorter-lived. These descriptors correlate with Bayes error, mean absolute residuals, and overconfidence. Our results suggest that zigzag persistence diagnoses the reliability of the inferred in-context task geometry and provides a context-level view of when TabPFN operates in topologically stressed regimes.

James Hu, Mahdi Ghelichi

Model Development Innovation, Risk Management TD Bank, Toronto, Canada

Tabular Foundation Models Neural Representation Geometry Tabular ICL

Jul 16, 2026cs.CV

Quantifying Training Membership Information in the Hyperspherical Embedding Geometry of Face Recognition Models

Face recognition models represent each face as an embedding vector on the unit hypersphere by clustering embeddings of the same identity while pushing different identities apart through angular-margin losses. Because these losses act only on training identities, non-member identities may form clusters with different geometric properties. In this paper, we quantify the magnitude of this difference and what training-time factors control it. We compute four statistics based on cluster geometry across 180 face recognition models in a factorial design over IResNet backbone size, loss head, training duration, and the number of training identities, and evaluate each configuration on nine benchmarks. Our results indicate that the number of training identities has the largest effect on member/non-member separability, while backbone and loss head contribute far less, and that, on a same-domain held-out reference, the geometric membership signal decreases monotonically as more identities are added to training. We provide an analysis of cross-domain (pose, age, quality, ethnicity) non-member benchmarks and report that these inflate the apparent membership signal. Finally, we fuse all four statistics with a learned classifier to reveal additional membership information beyond the best individual statistic.

Ünsal Öztürk, Sébastien Marcel

Idiap Research Institute, Martigny, Switzerland · Universit´e de Lausanne (UNIL), Lausanne, Switzerland

Neural Representation Geometry Membership Inference Attacks Face Recognition

Jul 13, 2026cs.LG

From Geometric Recovery to Causal Validation: A Reproducible Audit of Sparse Autoencoder Features, from Superposition Geometry to Causal Inertness

Sparse autoencoders (SAEs) are the standard for decomposing superposed neural representations into interpretable features, and evaluation relies predominantly on correlational recovery metrics -- cosine similarity between ground-truth directions and decoder atoms. We show this conflates two distinct claims: decoder-geometry alignment and encoder-activation behavior. We reproduce the superposition phase diagram of Elhage et al. (2022), identifying a convergence artifact at high sparsity and an under-described diffuse sharing regime at extreme overcompleteness. We reproduce the TopK-versus-L1 comparison of Gao et al. (2024), with direct evidence of L1 shrinkage. Our central result is causal: subjecting every recovered feature to ablation and steering, we find up to 77% of features passing a recovery bar (cosine >= 0.90) in a degraded SAE -- and 9% in a well-trained one -- are causally inert: the matched atom never fires when the feature is present, including matches at cosine ~1.000. We package the method as sae-causal-audit, a model-agnostic instrument with a deterministic pipeline. Re-auditing refines the finding: inertness decomposes by cause into structural inertness (antipodal-pair geometry, present in good SAEs) and competitive inertness (a TopK pathology of degraded SAEs), and by direction into read- and write-inertness, which five antipodal pairs dissociate completely -- unmonitorable yet steerable through the same atom, with steering specificities of 143-310 attached to zero ablation effects. We document why byte-exact reproducibility is unavailable by construction, and propose reporting it as a stack of claims with explicit scopes. Applying the instrument to a production SAE reproduces the pattern at small scale (14% inert) and surfaces an atom-collision signal: a handful of atoms recur as the nearest match for dozens of unrelated concepts, replicated across three batches.

Mohamed Abdessalem Bal

Independent Researcher, Algiers, Algeria

Feature Superposition Neural Representation Geometry Sparse Autoencoders Causal Interpretability

Jul 13, 2026cs.LG

How to Tame Grokking: Representation Geometry as a Control Signal

Grokking is a phenomenon in which neural networks initially memorize training data and only later exhibit strong generalization after prolonged optimization. Despite extensive recent study, the factors influencing the emergence and timing of grokking remain incompletely understood. We investigate the relationship between representation geometry and delayed generalization. We find that dimensionality collapse consistently precedes the onset of grokking in all evaluated settings. Motivated by these observations, we introduce Geometric Dimensionality Regularization (GeomDR), a simple spectral regularizer that modifies the effective dimensionality of hidden representations during training. Across modular addition, modular division, and permutation composition tasks, GeomDR consistently alters grokking dynamics and can substantially accelerate the onset of generalization depending on the intervention schedule and target dimensionality. In several settings, grokking is accelerated by up to 52 times relative to standard AdamW training. Similar qualitative effects are observed in both multilayer perceptrons and transformers. Together, these results suggest that representation geometry can serve as an effective control signal for grokking and provide evidence that geometric interventions offer a practical approach for studying and influencing delayed generalization in neural networks.

Maksim A Kazanskii

Independent Researcher

Neural Network Generalization Spectral Regularization Representation Collapse Neural Representation Geometry Neural Network Training Dynamics Grokking

Jul 12, 2026cs.LG

Beyond Coordinate Gauge: An Audited Protocol for Detecting Donor-Specific Functional Fingerprints after Neural Collapse

Independently trained neural networks have no shared neuron-index reference frame, so comparing them requires accounting for coordinate freedom. Neural Collapse sharpens this problem: networks converge toward a shared, low-dimensional geometry, raising the question of whether trajectory-specific functional variation remains distinguishable after convergence. We distinguish three claims - detectability, transplantability, and causal persistence - and address the first. Using five independently trained networks reconstructing Neural Collapse on MNIST, we apply a verified affine-correct alignment mapping donor heads into recipient coordinates. Donor-specific functional fingerprints remain distinguishable after recipient-level baseline correction: all 20 ordered donor-recipient pairs are correctly identified, with an exact permutation p=0.0083, robust to a leakage audit. These findings establish detectability under the test used here, but not transplantability or causal persistence. The study shows how alignment, ambiguity diagnostics, and leakage control combine to test cross-network variation in a controlled setting; whether this generalizes beyond it is open.

Truong Xuan Khanh, Phan Thanh Duc

H&K Research Studio, Clevix LLC, Hanoi, Vietnam · Banking Academy of Vietnam, Hanoi, Vietnam

Neural Representation Geometry Neural Collapse

Jul 9, 2026cs.LG

Contravariance Theory: Strong Alignment for Minimal Solutions to Hard Tasks

A series of results from the NeuroAI over the past fifteen years have raised core questions both about how to compare Deep Neural Network (DNN) models to the brain, and about how much convergent evolution to expect between artificial networks and real brain networks. Here, we show that for any two minimal DNN solutions to a sufficiently hard task: (i) "weak" alignment of network representations based on affine mappings guarantees "strong" alignment of privileged axes, and (ii) alignment "zippers" up the network hierarchy, causing the emergence of privileged axes from end-to-end task optimization. These results formalize the notion of contravariance from Cao and Yamins [2024], and illustrate important consequences for the theory of NeuroAI: with sufficiently strong tasks, choice of metric for inter-network comparison is not all that sensitive, and that convergent evolution is probably inevitable.

Dan Yamins, Aran Nayebi

Departments of Computer Science and Psychology, and Wu Tsai Neurosciences Institute, Stanford University, Stanford, CA 94305 USA · Machine Learning Department, Neuroscience Institute, and Robotics Institute, Carnegie Mellon University, Pittsburgh, PA 15213 USA

Neural Representation Geometry

Jul 8, 2026cs.NE

Dynamic neural manifolds for flexible closed-loop control on neuromorphic hardware

In biological circuits, sequential neural activity evolves along dynamic, low-dimensional manifolds to enable flexible behavior. Spiking network models link aspects of this sequential activity to features of manifold geometry through specific circuit mechanisms, making dynamic neural manifolds parameterizable, and thereby offering an explainable framework for neural computation. Extending this framework to neuromorphic engineering, we present an implementation on the SpiNNaker 2 chip for real-time, closed-loop control. By allowing sensory inputs to modulate heterogeneous inhibition, gain, and transient currents, our architecture drives rapid subspace rotations to switch between behaviors, as well as fine-grained trajectory control within them. We validate this via a robotic simulation where an agent uses sensory feedback to dynamically reconfigure its manifold geometry to navigate through a maze. Our results establish dynamic manifolds as a feasible approach for explainable neuromorphic architectures and a substrate for investigating biological neural dynamics.

Oskar von Seeler, Christian Tetzlaff, Andrew Lehr

Group of Computational Synaptic Physiology, Department of Neuro- and Sensory Physiology, University Medical Center Göttingen, Germany · Campus Institute Data Science, University of Göttingen, Germany · Circulant Labs, Bensheim, Germany

Robotic Control Neural Representation Geometry Spiking Neural Networks Neuromorphic Computing

Jul 5, 2026cs.CL

Language Models Represent and Transform Concepts with Shared Geometry

How concepts are represented in neural networks is a fundamental question in machine learning. The dominant view treats concept representations as stationary geometric objects. Yet concepts appear in context, and context transforms them. Drawing from neural population geometry, we formalize concept representations as point-cloud manifolds and contextual transformations as vector fields, and instantiate this framework in large language models. Across six model families of varying scales, we find that context moves each concept differently. The variance in these displacements is semantically organized, correlating with lexical concreteness and density. Importantly, both the concepts being transformed and this variance structure are shared across models: displacement structure transported from one model predicts held-out displacements in others significantly above chance. Together, these findings show that models share a common geometry not only in how concepts are represented, but more importantly in how context transforms them, a structure with richer organization than prior work has recognized.

Zhimin Hu, Lanhao Niu, Sashank Varma

1Georgia Institute of Technology · University of Edinburgh

Neural Representation Geometry Representation Geometry in Language Models

Jul 3, 2026cs.LG

Transition Information Density: Morphological Trajectories, Synesthetic Perception, and Structured Interpolation in Neural Training (or: The Synesthetic AI)

Standard machine learning training presents data as discrete endpoint pairs, omitting the structure of the space between them. This paper introduces Transition Information Density (TID) -- the information content recoverable from structured intermediate states between categorically distinct training endpoints -- and Positional Identity, the defined location of an intermediate state on the A-to-B continuum. Both constructs are grounded in three empirical contexts: grapheme-color synesthesia, the Synesthesia Grid (a boundary-contour morphing algorithm instantiating TID in visual morphological space), and a four-condition training experiment across four representational mediums. Probes trained on structured interpolation at defined Positional Identities (C3) exhibit substantially lower intrinsic dimensionality than volume-matched controls (C2) in Phonetic/Linguistic (C3: 3.33 vs. C2: 10.81) and Semantic Description (C3: 4.59 vs. C2: 8.67) mediums. Visual and cross-modal mediums do not show this effect, establishing a modality boundary condition. A fixed-N=50 comparison confirms that Positional Identity structure, not sample count, drives the effect. Resolution N scales monotonically with representational richness. Pooled TwoNN analysis reveals globally collapsed representations in visual space (0.075) and globally consistent representations in phonetic space (0.977). The paper contributes a formal definition of TID and Positional Identity, a nine-metric shape characterization framework, and a four-condition experimental design isolating trajectory structure, data volume, and Positional Identity as distinct factors.

Sam Mao

New York University | Interactive Media Arts

Neural Representation Geometry Trajectory Representation Learning

Jul 3, 2026cs.LG

Observable- and Positional-Encoding-Dependent Symmetry Readout from Neural Network Weights

Post-hoc analysis of trained neural network weights often seeks to recover geometric structure directly from the parameters. We show that, for positional-encoding-equipped neural fields, the symmetry visible from weights is not the true symmetry group itself, but an observable symmetry set determined by the trained parameters, the positional encoding (PE), and readout observable. We formulate this dependence through an exact observability hierarchy,

G_{\mathrm{obs}}^{\mathrm{exact}} \subseteq G_{\mathrm{lift}}^{\mathrm{exact}}(φ) \cap G_{\mathrm{true}}

, where

G_{\mathrm{lift}}^{\mathrm{exact}}(φ)

is the set of input transformations that the PE can exactly lift to the feature space. The hierarchy implies that even when a target function has a geometric symmetry, that symmetry may be structurally invisible to weight-level observables if the PE does not represent the corresponding transformation. We test this prediction using MLPs trained on two-dimensional signed distance functions with multiple shape symmetry groups, positional encodings, and Gram-based observables. The results show a consistent PE-dependent pattern: DyadicAxisPE supports

D_4

-sensitive readout but structurally suppresses

D_3

rotations, TriAxisPE yields lower

D_3

/

D_6

readout scores under the tested Gram observables by replacing coordinate axes with three 120-degree-separated axes, and random Fourier features mainly exhibit a

π

-rotation response under these readouts. These findings show that PE design affects not only approximation behavior but also which structures are accessible to post-hoc weight-level readouts. This provides a basis for a principled observable-dependent symmetry readout.

Naoya Chiba, Satoshi Sugiyama, Yuki Uranishi

D3 Center, The Unversity of Osaka

Neural Representation Geometry Implicit Neural Representations Positional Encoding Weight-Space Analysis

Jul 2, 2026cs.CV

Transformer Geometry Observatory TGO-II: Representational Similarity Observatory

While Vision Transformers have achieved remarkable success across computer vision and language applications, the geometric evolution of their internal representations throughout training remains insufficiently understood. Existing analyses primarily focus on attention mechanisms and downstream performance, leaving the evolution of representation geometry largely unexplored. In this work, we present Transformer Geometry Observatory-II (TGO-II), a representation geometry analysis framework designed to investigate how Transformer representations evolve during supervised training. TGO-II analyzes Vision Transformer (ViT-Small/16) representations using Centered Kernel Alignment (CKA), Singular Vector Canonical Correlation Analysis (SVCCA), Two-Nearest Neighbor Intrinsic Dimensionality (TwoNN-ID), and token covariance analysis. Our experiments reveal three key observations. First, both CKA and SVCCA progressively decrease throughout training, indicating increasing representational specialization across Transformer layers. Second, intrinsic dimensionality consistently increases before stabilizing, suggesting progressive expansion of the representation manifold into a larger set of locally accessible degrees of freedom. Third, token covariance and coupling analyses demonstrate that strong token interaction structure persists throughout training, challenging the hypothesis that increasing representational complexity arises primarily from progressive token independence. These findings suggest that representation complexity and layer specialization emerge simultaneously during training. Manifold expansion appears to occur without token decoupling. Together, these observations motivate a new hypothesis in which Vision Transformers increase representational complexity through progressively richer transformations while preserving strong token interaction structure during learning.

Kaustubh Kapil, Kishor P. Upla

Department of Electronics Engineering Sardar Vallabhai National Institute of Technology (SVNIT), Surat, India

Intrinsic Dimensionality Representational Similarity Analysis Neural Representation Geometry Visual Representation Learning Vision Transformer

Jul 2, 2026cs.CV

Understanding Geometric Representations in Self-Supervised Vision Transformers via Subspace Intervention

We introduce a controlled subspace intervention framework to investigate how self-supervised Vision Transformers (ViTs) encode dense geometric information. While linear probing is widely used to assess geometric representations, it treats features as a black box, failing to disentangle the underlying topology. To address this issue, we decompose the weights of converged linear probes to isolate the low-rank subspaces containing explicit geometric signals using Singular Value Decomposition (SVD). Our perspective yields three key insights: (1) Pre-training objectives determine how features are encoded. DINOv2 aligns spatial features for efficient linear extraction, while Masked Autoencoders (MAE) tend to disperse these signals, requiring a broader spatial context. (2) Explicit geometric representations are highly compressible, suggesting dense predictive heads could potentially be constrained to low-rank subspaces with minimal performance loss. (3) The layer-wise task affinity suggests that geometric precision peaks at intermediate layers before yielding to semantic abstraction in the final layers. By connecting internal encoding mechanics with downstream performance, these findings provide a basis for effective feature selection and lightweight decoder design. The source code is available at https://github.com/Zhou-Weichen/Geosubprobe.

Weichen Zhou, Yawen Zou, Chunzhi Gu +3

University of Toyama, Toyama, Japan · University of Fukui, Fukui, Japan · Chukyo University, Aichi, Japan +1

Neural Representation Geometry Linear Probing Vision Transformer Geometric Representation Learning Self-Supervised Visual Representation Learning

Jul 2, 2026cs.LG

Finite-Lag Operator Geometry of Recurrent Representations

Recurrent representations are trajectories, but representation geometry is often measured from static snapshots. We develop finite-lag operator geometry for recurrent hidden states from observed source-successor pairs

(X_t,X_{t+Δ})

. The primitive is the conditional transport law

Q_Δ(dy\mid x)

, estimated by a dense Gaussian source-smoothing operator. From this directed finite-lag law we derive a source-centered transport tensor

G_Δ

, which decomposes exactly into conditional spread and coherent displacement, and an antisymmetric coordinate circulation

W_Δ^ρ

, which summarizes directed lagged flow. We prove affine covariance with explicit metric dependence of scalar summaries, dense estimator stability on bounded trajectory clouds, and a finite-lag separation result showing that source-centered transport detects deterministic recurrent motion not recorded by infinitesimal carre-du-champ geometry. A linear-Gaussian closed form calibrates the quantities in terms of the update

A_Δ

, source covariance, and innovation covariance. Controlled experiments validate the decomposition, circulation, covariance, and stability predictions. In performance matched repeat-copy networks, the framework reveals architecture dependent differences in total transport scale and coherent displacement trace, while coherent displacement fraction is metric and resolution dependent.

Kanishka Reddy

Department of Applied Mathematics University of Washington Seattle, WA 98105

Neural Representation Geometry Recurrent Neural Networks Geometric Representation Learning

Jun 30, 2026cs.LG

Radial Suppression Accelerates Algorithmic Generalization: A Geometric Analysis of Delayed Generalization

Why do neural networks memorize algorithmic training data long before they generalize? We present a geometric case study demonstrating that, on tasks where generalization requires discovering structured low-dimensional circuits, the memorization-generalization delay is driven by radial inflation of hidden representations under cross-entropy optimization. We formalize a radial-angular decomposition of activation-space dynamics and derive three testable propositions: (i) that penalizing radial inflation induces anisotropic, data-dependent weight regularization; (ii) that it suppresses radial gradient energy below the isotropic random baseline, forcing predominantly angular updates; and (iii) that it biases convergence toward flatter minima. To empirically validate these propositions, we study a single-hyperparameter norm penalty that softly constrains activations to a sqrt(d)-radius hypersphere. On modular arithmetic, this penalty accelerates grokking up to 6x across MLPs and Transformers, and halves training steps for a 10M-parameter nanoGPT on 3-digit addition.

Srijan Tiwari, Aditya Chauhan, Manjot Singh

Indian Institute of Technology Roorkee

Neural Network Generalization Neural Representation Geometry Neural Network Memorization Neural Network Training Dynamics Feedforward Neural Networks Grokking

Jun 30, 2026cs.LG

Resolving superposition in AI for interpretability and cross-modal alignment in patient-neuronal images

Artificial intelligence is transforming our capability to solve biological challenges. In dimensionality bottleneck regimes exacerbated by high-dimensional biological data, neural networks force distinct concepts into the lower dimensions known as superposition. Although this superposition is widely known to hinder interpretability, its impact on corrupting the geometry of latent spaces remains critically overlooked. Here, we utilized sparse autoencoders (SAEs) trained on over 100,000 multiplexed images of patient-derived Parkinson's disease and healthy neurons to resolve superposition. This approach bypasses the mathematical non-uniqueness of feature attribution by shifting to interpretable latent representation analysis. We theoretically and empirically demonstrate that superposition contaminates representational metric spaces, and thereby SAEs successfully recover geometric fidelity. By treating these geometrically purified representations as single-cell state vectors, we adapted single-cell RNA sequencing (scRNA-seq) data analysis methodologies directly to the image domain. Finally, we introduce GW-map, utilizing Gromov-Wasserstein optimal transport to align these image representations with authentic scRNA-seq data de novo. This coupling reconstructs hierarchical neuronal pathology pathways such as Calcium-AIS scaffold, without reference spatial transcriptomics, establishing a scalable foundation for spatial biology. Code is available at https://github.com/jijihihi/Bio\_superposition

Jisung Park, Seohyeon Kang, Daeun Yoo +8

1KAIST · 2Konyang University · 4UCL Queen Square Institute of Neurology & The Francis Crick Institute +1

Feature Superposition Neural Representation Geometry Cross-Modal Alignment Representation Disentanglement Sparse Autoencoders Microscopy Image Analysis

Jun 29, 2026cs.LG

Learning as Observable Matrix Dynamics: Diffusive Relaxations versus Phase Transitions

Observable Matrix Dynamics (OMD) is a diagnostic framework that probes the dynamics of high-dimensional internal representations of inputs by a neural network via a fixed-size

N \times N

distance matrix

M(t)

on a held set of

N

inputs. OMD uses methods of random matrix theory and particle dynamics to explore spectral reorganisations that are missed by scalar loss functions, but are informative of the training process. We read

M(t)

against a perturbative ambient-versus-latent decomposition extending the Bogomolny--Bohigas--Schmit (BBS) theory of random distance matrices, with per-snapshot diagnostics for the top-of-spectrum band structure and ambient noise, trajectory-level observables linking snapshots, and a 3D MDS embedding (bottom-three eigenvectors) rendering training as a moving particle cloud. Across seven experiments, diffusive regimes lack stable top-of-spectrum band structure, while sharp endogenous or externally driven reorganisations produce stable fingerprints: consistent with smooth or product latent geometries in BBS-adjacent cases, and with finite-cluster or Fourier-soliton structures otherwise. OMD thus reads the geometric regime of a representation rather than reporting a single intrinsic dimension.

Igor Halperin

All calculations, numerical analysis, and manuscript preparation were performed by Claude Code with Opus 4.7 working as an AI assistant under author's supervision.

Neural Representation Geometry Neural Network Training Dynamics

Jun 28, 2026q-bio.NC

Geometric Reliability of Neural Population Codes: Sampling Calibration and Within-Session Nonstationarity

Trial-to-trial variability limits how reliably neural population geometry can be estimated, while comparisons across populations depend on neuron and trial counts, response quality, and clustered sampling. We quantified within-session geometric reliability using Shesha, the Spearman correlation between representational dissimilarity matrices estimated from independent trial subsets, in all 39 Steinmetz Neuropixels sessions and in olfactory bulb and piriform cortex recordings from Bolding and Franks. Steinmetz analyses matched neurons and repetitions, compared observed reliability with a stationary residual-bootstrap expectation, and used mouse-level or mouse-clustered inference. Mean matched reliability was 0.0402 across 312 area-by-session recordings. Regional differences and reliability above the stationary benchmark did not survive correction. Temporal effects received the strongest support: interleaving early and late trials increased reliability relative to blocked allocation (

Δ=0.02666

,

q=0.001953

), and RDM similarity declined with within-session lag (mean mouse-level slope

=-0.01912

,

q=0.001953

;

n=10

mice). Outer-cross-fitted reliability was not associated with choice-direction coupling or stimulus or response-direction decoding after correction. Olfactory comparisons remained descriptive because few paired sessions and no animal identities were available. In held-out simulations, associative recurrence outperformed feedforward subspace denoising but not divisive normalization. Representational geometry became less reproducible with temporal separation within a session, and comparisons across neural populations require sampling calibration and independent inference.

Prashant C. Raju

Representational Similarity Analysis Neural Representation Geometry Temporal Consistency

Jun 28, 2026cs.CL

Anisotropy Decides Cosine vs. Rank Metrics for Text Embeddings

The standard way to compare two text embeddings is cosine similarity. Scattered studies report that a different metric does better, but never pin down the geometric condition that decides when, or why. We settle both with a comprehensive empirical study: nineteen parameter-free similarity metrics on nineteen encoders, from compact sentence transformers up to seven-billion-parameter large language models, across seven datasets. The answer is geometric. When an encoder spreads its variance evenly across directions, cosine is the best parameter-free choice and no other metric helps by a usable margin. When the variance concentrates into a few dominant directions, a property known as anisotropy, rank-based and L1-type metrics beat cosine by a clear margin. The absolute gain is modest, but because cosine starts low on these encoders it is a sizable relative improvement, around twenty percent on average and largest where cosine is weakest. What decides this is the geometry of the embedding space, not how the model was trained: where the two disagree, the metric follows the geometry. One number, the fraction of variance held by the single most dominant dimension, predicts how much the alternatives help across all nineteen encoders, with a rank correlation of 0.86 and a linear correlation of 0.95. To test this as the cause rather than a correlate, we project out the dominant directions: cosine recovers and the advantage of the other metrics nearly vanishes, but only on the encoders that were anisotropic to begin with. The effect is directional, not magnitude based, since it survives normalizing every vector to unit length. Among parameter-free metrics, then, cosine is the right tool wherever an encoder is well spread, which includes the fine-tuned embedders commonly deployed for retrieval, and we give a one-number diagnostic for when it is not.

V. S. Raghu Parupudi

University of California, San Diego

Semantic Textual Similarity Neural Representation Geometry Text Embedding Evaluation Representation Geometry in Language Models

Jun 26, 2026cs.LG

Geometric Measurements of the Axiom of Choice in Neural Proof Embeddings

The axiom of choice has divided the foundations of mathematics for over a century, but the distinction between classical and constructive proofs has remained a philosophical and methodological one. We use Lean 4's kernel-level tracking of axiom dependence to show that the axiom of choice has a measurable geometric correlate in proof space that obeys a one-parameter mixture law and has operational consequences for neural theorem provers. To do this, we partition

471{,}260

declarations of Mathlib by transitive dependence on the axiom of choice and represent a filtered population of

42{,}355

traced theorems by their sequences of tactic invocations. We use the constructive proofs in this dataset to train a self-supervised proof encoder and show that when using it to measure classical proofs, three complementary measurements (anomaly score, reconstruction loss, and density-superlevel containment) exhibit a common decline with the proof's distance from the axiom in the dependency graph, from sharp separation at the shallow boundary (AUC

0.847

at distance

2

) to indistinguishability at distance~

9{+}

. Robustness controls show that the signature survives length, file, author, and topic controls, and replicates under full-source encoders trained on normalised proof source. Operationally, we show that on an evaluation sample of

251

Mathlib theorems, Lean's \texttt{aesop} tactic solves constructive theorems at

13\times

the rate of classical ones, and a neural-guided hybrid using the ReProver tactic generator compresses the gap to

5\times

. The geometric anomaly score predicts \texttt{aesop} failure beyond proof length, providing an operational link between the geometric signature and prover performance.

Rodrigo Mendoza-Smith

Independent Researcher.

Neural Representation Geometry Automated Theorem Proving Lean Theorem Proving

Jun 26, 2026q-bio.NC

CANNs: A Toolkit for Research on Continuous Attractor Neural Networks

Continuous attractor neural networks (CANNs) are the canonical computational framework for how the brain encodes continuous variables such as spatial position, head direction, and movement direction, and explain the activity of hippocampal place cells, entorhinal grid cells, and head-direction cells. CANN research, however, is fragmented: most results rest on lab-specific implementations, general-purpose simulators lack CANN-specific abstractions, and the path from spike trains to attractor geometry in real recordings lacks a standardized toolkit. Here, we present a comprehensive open-source toolkit that unifies the full CANN research workflow. It combines three tightly integrated components: 1) canns, a Python library on BrainPy/JAX that provides standardized 1D/2D CANNs, spike-frequency-adaptation variants, grid cell networks, hierarchical path-integration models, and brain-inspired attractor architectures, together with curated datasets, task generators, an analyzer module and trainer modules for biologically plausible plasticity; 2) canns-lib, a Rust acceleration backend delivering hundreds-of-times speedups for spatial-navigation workloads and modest gains for Ripser-based persistent homology; 3) ASA (Attractor Structure Analyzer), a PySide6 pipeline applying persistent homology and cohomology to experimental neural recordings to detect ring-like and toroidal attractor signatures in real data. The toolkit ships with full-detail reproducible pipelines that recover recent CANN results including SFA-driven anticipative tracking, theta sweeps in head-direction/place/grid systems, and hierarchical path integration.

Sichao He, Aiersi Tuerhong, Shangjun She +4

School of Psychological and Cognitive Sciences, Peking–Tsinghua Center for Life Sciences, PKU-IDG/McGovern Institute for Brain Research, Center of Quantitative Biology, Peking University, Beijing, China · College of Mathematics and Statistics, Chongqing University, Chongqing, China

Neural Representation Geometry Recurrent Neural Networks Neural Encoding

Jun 25, 2026cs.CV

SubdivAR: Autoregressive Next-Scale Prediction for Neural Mesh Subdivision

Mesh subdivision is a fundamental operation for converting coarse, editable meshes into high-resolution surfaces, with broad applications in digital asset creation. Classical rule-based schemes rely on fixed local refinement rules and often produce over-smoothed surfaces. Recent neural subdivision methods improve detail synthesis, but remain constrained by local modeling and exhibit limited generalizability. We present SubdivAR, a neural mesh subdivision framework based on our proposed Mesh Autoregressive Representation (MAR). MAR arranges meshes at different subdivision levels into an ordered scale sequence, reformulating subdivision as autoregressive next-scale prediction. To support this formulation, we introduce a Hybrid Topology-Aware Transformer that combines global semantic attention with topology-constrained local feature aggregation. SubdivAR adopts a next-scale coordinate prediction paradigm, regressing vertex offsets at each refinement stage to preserve subdivision topology while recovering fine-grained geometric details. To enable reliable learning, we construct FII-40K, a curated dataset of nearly 40,000 high-quality meshes with multi-level subdivision supervision. Experiments show that SubdivAR outperforms state-of-the-art baselines, reducing Hausdorff Distance and Chamfer Distance by 18.8% and 14.2%, respectively, and demonstrates strong robustness on complex open-surface geometries.

Huipeng Guo, Zikai Song, Hang Long +7

Huazhong University of Science and Technology

3D Mesh Generation Neural Representation Geometry Transformer Autoregressive Generation

Jun 24, 2026cs.RO

Exploring the Intrinsic Geometry of Diffusion Models with Constrained Inverse Kinematics

Recent studies suggest that diffusion models can recover geometric structure in the data manifolds they are trained on, yet the supporting evidence has so far come mostly from natural-image data, where the underlying geometry itself is unknown. We study this question in a setting where the geometry is analytically tractable: constrained inverse kinematics (IK). Each task-space constraint defines a configuration-space manifold with known intrinsic dimension, giving direct ground truth for evaluating the geometry learned by the model. For each of the 6-DoF UR5 and 7-DoF Franka, we train a single conditional diffusion model across seven constraint families, spanning solution manifolds from discrete IK branches to self-motion manifolds. Our empirical results reveal that the intrinsic dimension recovered from the model's score function matches the analytical degrees of freedom of the corresponding constraint manifold across both robots. Moreover, linear interpolation in the latent space leads to generated solutions that remain close to the appropriate constraint manifold, indicating that the learned representation further captures geometric structure of the constraint family beyond intrinsic dimension alone. Constrained IK therefore offers a controlled setting for studying the intrinsic geometry learned by diffusion models.

Miguel Angel Rogel Garcia, Phone Thiha Kyaw, Jonathan Kelly

Institute for Aerospace Studies, University of Toronto, Toronto, Canada

Neural Representation Geometry Conditional Diffusion Models Diffusion Models

Jun 23, 2026cs.CV

Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds

What is the geometry of a visual percept? The most widely used protocols for decomposing neural network representations into interpretable parts treat concepts as isolated directions, yet recent work shows that concepts are often realized as geometric structures in low dimensional regions of activation space. We turn to the literature of Structured sparsity to close this gap, and show that block sparsity, which groups directions into blocks, is the prior matched to a generative model in which a representation is a sparse sum of low-dimensional manifolds: the modern, learned form of a classical idea in visual neuroscience, where a visual feature is carried by a coordinated group of neurons rather than a single tuned one. We implement three variants of block-sparse featurizers (BSFs) and, through a minimum-description-length analysis, show that all three describe activations more compactly than direction-based featurizers, with the recovered concepts typically two- to four-dimensional. We then use BSFs to (i) recontextualize prior work, showing that curve detectors in InceptionV1 actually read from a single continuous curve manifold, (ii) discover novel manifolds including shadows and lighting in DINOv3, and (iii) support interpretable control of image generation in diffusion models (SDXL) via manifold steering.

Thomas Fel, Matthew Kowal, Mozes Jacobs +22

•Goodfire · aHarvard University · bStanford University +1

Neural Representation Geometry Activation Sparsity Structured Sparsity Neural Network Interpretability Manifold Learning