Rethink MAE with Linear Time-Invariant Dynamics

• URL: http://arxiv.org/abs/2605.00915v1
• Date: Wed, 29 Apr 2026 15:07:00 GMT
• Title: Rethink MAE with Linear Time-Invariant Dynamics
• Abstract summary: We show that token order is a critical, exploitable dimension in frozen visual representations.<n>We propose SSMProbe, a probing framework driven by a State Space Model.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Standard representation probing for visual models relies on mathematically permutation-invariant operations like Global Average Pooling (GAP) or CLS tokens, treating patch representations as an unstructured bag-of-words. We challenge this paradigm by demonstrating that token order is a critical, exploitable dimension in frozen visual representations (e.g., MAE, BEiT, DINOv2, and ViT as CLS-ablation extreme). We propose SSMProbe, a probing framework driven by a State Space Model (SSM). Operating as discrete Linear Time-Invariant (LTI) dynamical systems, SSMs act as permutation-sensitive probes where sequence order strictly dictates the final state due to inherent memory decay. Formulating token ordering as an information scheduling problem, we compare fixed scan heuristics against a differentiable soft permutation (Sinkhorn-based) learned from downstream supervision. Evaluations on standard and fine-grained classification benchmarks reveal a striking order gap: while fixed scans fail dramatically on highly localized patch features, our learned soft permutation successfully extracts highly competitive performance from otherwise heavily localized patch sequences. We find that pre-training objectives fundamentally shape token structure: DINOv2 concentrates global semantics in optimized CLS tokens leaving patches hyperspecialized, pure MAE preserves distributed representations with heterogeneous patch informativeness, and ViT represents a supervised CLS-dominated extreme. BEiT occupies middle ground. This heterogeneity is order-dependent -- meaning the SSM probe's performance depends critically on which tokens are placed at which temporal positions -- and is not merely a topological property of the spatial grid. SSMProbe's learned routing effectively discovers and exploits this heterogeneity, offering a powerful new diagnostic lens for visual representation analysis.

Related papers

• Where Does Generative Difficulty Reside? An Empirical Study of Target Representations [62.54876287800644]
  The target representation defines the distribution an image generator must learn, yet it is often treated as an interchangeable interface.<n>We study raw pixels, SD-VAE latents and DINOv2 as well as MAE representation-autoencoder features within a unified masked autoregressive rectified-flow model.<n>Our results show that compression, reconstruction fidelity, token dimensionality, and visible semantic clustering do not individually predict generative behavior.
  arXiv  Detail & Related papers  (2026-08-01T12:32:31Z)
• ZACH-ViT: Regime-Dependent Inductive Bias in Compact Vision Transformers for Medical Imaging [0.0]
  ZACH-ViT is a compact Vision Transformer that removes both positional embeddings and the [token] token.<n>"Zero-token" specifically refers to removing the dedicated [NIST] aggregation token and positional embeddings.<n>ZACH-ViT achieves competitive performance while maintaining sub-second inference times.
  arXiv  Detail & Related papers  (2026-02-20T01:38:59Z)
• SMKC: Sketch Based Kernel Correlation Images for Variable Cardinality Time Series Anomaly Detection [0.0]
  In operational environments, monitoring systems frequently experience sensor churn.<n>We propose SMKC, a framework that decouples the dynamic input structure from the anomaly detector.<n>We find that a detector using random projections and nearest neighbors on the SMKC representation performs competitively with fully trained baselines.
  arXiv  Detail & Related papers  (2026-01-28T21:15:11Z)
• Improving Semantic Uncertainty Quantification in LVLMs with Semantic Gaussian Processes [60.75226150503949]
  We propose a Bayesian framework that quantifies semantic uncertainty by analyzing the geometric structure of answer embeddings.<n>S GPU maps generated answers into a dense semantic space, computes the Gram matrix of their semantic embeddings, and summarizes their semantic configuration.<n>We show that S GPU transfers across models and modalities, indicating that its spectral representation captures general patterns of semantic uncertainty.
  arXiv  Detail & Related papers  (2025-12-16T08:15:24Z)
• Spanning Tree Autoregressive Visual Generation [51.7635842702602]
  We present Spanning Tree Autoregressive (STAR) modeling, which can incorporate prior knowledge of images, such as center bias and locality, to maintain sampling performance.
  arXiv  Detail & Related papers  (2025-11-21T09:45:17Z)
• Robust Canonicalization through Bootstrapped Data Re-Alignment [5.437226012505534]
  Fine-grained visual classification tasks, such as insect and bird identification, demand sensitivity to subtle visual cues.<n>We propose a bootstrapping algorithm that iteratively re-aligns training samples by reducing variance.<n>We show that our method consistently outperforms equivariant, and canonicalization baselines while performing on par with augmentation.
  arXiv  Detail & Related papers  (2025-10-09T13:05:20Z)
• Structured Sparse Transition Matrices to Enable State Tracking in State-Space Models [68.31088463716269]
  We propose a structured sparse parametrization of transition matrices in state-space models (SSMs)<n>Our method, PD-SSM, parametrizes the transition matrix as the product of a column one-hot matrix ($P$) and a complex-valued diagonal matrix ($D$)<n>The model significantly outperforms a wide collection of modern SSM variants on various FSA state tracking tasks.
  arXiv  Detail & Related papers  (2025-09-26T12:46:30Z)
• SemaMIL: Semantic-Aware Multiple Instance Learning with Retrieval-Guided State Space Modeling for Whole Slide Images [17.674866281320046]
  SemaMIL is an adaptive method for extracting discriminative features from whole slide images.<n>It clusters semantically similar patches in sequence through a reversible permutation.<n>It achieves state-of-the-art subtype accuracy with fewer FLOPs and parameters.
  arXiv  Detail & Related papers  (2025-08-30T10:13:18Z)
• On the Expressiveness and Length Generalization of Selective State-Space Models on Regular Languages [56.22289522687125]
  Selective state-space models (SSMs) are an emerging alternative to the Transformer.<n>We analyze their expressiveness and length generalization performance on regular language tasks.<n>We introduce the Selective Dense State-Space Model (SD-SSM), the first selective SSM that exhibits perfect length generalization.
  arXiv  Detail & Related papers  (2024-12-26T20:53:04Z)
• Semantic Equitable Clustering: A Simple and Effective Strategy for Clustering Vision Tokens [57.37893387775829]
  We introduce a fast and balanced clustering method, named Semantic Equitable Clustering (SEC)<n>SEC clusters tokens based on their global semantic relevance in an efficient, straightforward manner.<n>We propose a versatile vision backbone, SECViT, to serve as a vision language connector.
  arXiv  Detail & Related papers  (2024-05-22T04:49:00Z)
• Quick Adaptive Ternary Segmentation: An Efficient Decoding Procedure For Hidden Markov Models [41.99844472131922]
  Decoding the original signal from the noisy observations is one of the main goals in nearly all HMM based data analyses.<n>We present QATS, a divide-and-conquer algorithm with computational polylogarithmic complexity in the length of the sequence, and cubic in the size of the state space.<n>An implementation of QATS is in the R-package QATS on GitHub.
  arXiv  Detail & Related papers  (2023-05-29T19:37:48Z)
• Performance Embeddings: A Similarity-based Approach to Automatic Performance Optimization [71.69092462147292]
  Performance embeddings enable knowledge transfer of performance tuning between applications. We demonstrate this transfer tuning approach on case studies in deep neural networks, dense and sparse linear algebra compositions, and numerical weather prediction stencils.
  arXiv  Detail & Related papers  (2023-03-14T15:51:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.