Neural Architecture Distributions: A New Paradigm for Stochastic Segmentation

• URL: http://arxiv.org/abs/2606.21061v1
• Date: Fri, 19 Jun 2026 03:14:00 GMT
• Title: Neural Architecture Distributions: A New Paradigm for Stochastic Segmentation
• Abstract summary: segmentation seeks to represent multiple plausible masks for a single image.<n>Most existing methods introduceity by injecting continuous latent variables or by iterative denoising trajectories.<n>We propose architecture distributions as a new source for segmentation.
• Score: 8.128553138697024
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Stochastic segmentation seeks to represent multiple plausible masks for a single image, which is essential in safety- and quality-critical applications such as medical imaging or building defect inspection. Most existing methods introduce stochasticity by injecting continuous latent variables or by iterative denoising trajectories, whose stochastic sources are difficult to search or audit directly. We propose architecture distributions as a new stochastic source for segmentation: instead of sampling a latent variable or noise, we sample a discrete architecture from a learned distribution over operator choices at multiple searchable positions in a segmentation backbone. Each sampled architecture yields one mask through the selected active path, so inference depends on the executed subnet rather than the complete candidate bank. This approach also supports architectural provenance, since each output corresponds to a specific architecture configuration. To reduce collapse toward averaged masks, we train with set-level supervision by matching a set of architecture-sampled predictions to the annotation set using an IoU-based energy-distance surrogate. We further construct the candidate bank with evolutionary search, making the support of the stochastic source optimizable before distribution learning. The proposed method achieves state-of-the-art distribution matching and hypothesis coverage on LIDC-IDRI, and remains effective on two extension tasks. To the best of our knowledge, this is the first work to formulate stochastic segmentation as learning an architecture distribution and realizing output diversity through architecture sampling.

Related papers

• Source-Agnostic Image Translation Based on Latent Aware Adaptive Masking [2.0673681624647844]
  We propose a source-agnostic framework that dynamically refines a binary mask throughout the reverse diffusion process.<n>Our method introduces a time-dependent statistical thresholding scheme derived from the empirical mean and standard deviation of prediction discrepancies.<n>Our approach enables precise, automated localization and seamless translation across diverse source distributions without any specialized training.
  arXiv  Detail & Related papers  (2026-08-14T07:51:40Z)
• Open-Set Source Tracing as Compositional Factors via Structured Prototypes [7.66732820915854]
  Current research often equates a "source" with its generative architecture.<n>We propose redefining a source as a compositional of Architecture, Training Data, and other training factors affecting the generated speech.<n>This approach improves performance for partially seen sources and maintains robustness in fully open-set scenarios.
  arXiv  Detail & Related papers  (2026-07-03T09:23:36Z)
• Discriminative Flow Matching Via Local Generative Predictors [0.0]
  We propose a framework that reformulates classification and object detection as a conditional transport process.<n>Our method attaches multiple independent flow predictors to a shared backbone.<n>This architecture provides the flexibility to update blocks either sequentially to minimise activation memory or in parallel to suit different hardware constraints.
  arXiv  Detail & Related papers  (2026-03-14T12:56:29Z)
• CSMCIR: CoT-Enhanced Symmetric Alignment with Memory Bank for Composed Image Retrieval [54.15776146365823]
  Composed Image Retrieval (CIR) enables users to search for target images using both a reference image and manipulation text.<n>We propose CSMCIR, a unified representation framework that achieves efficient query-target alignment through three synergistic components.
  arXiv  Detail & Related papers  (2026-01-07T09:21:38Z)
• A Refreshed Similarity-based Upsampler for Direct High-Ratio Feature Upsampling [54.05517338122698]
  A popular similarity-based feature upsampling pipeline has been proposed, which utilizes a high-resolution feature as guidance.<n>We propose an explicitly controllable query-key feature alignment from both semantic-aware and detail-aware perspectives.<n>We develop a fine-grained neighbor selection strategy on HR features, which is simple yet effective for alleviating mosaic artifacts.
  arXiv  Detail & Related papers  (2024-07-02T14:12:21Z)
• Dynamic Perceiver for Efficient Visual Recognition [87.08210214417309]
  We propose Dynamic Perceiver (Dyn-Perceiver) to decouple the feature extraction procedure and the early classification task. A feature branch serves to extract image features, while a classification branch processes a latent code assigned for classification tasks. Early exits are placed exclusively within the classification branch, thus eliminating the need for linear separability in low-level features.
  arXiv  Detail & Related papers  (2023-06-20T03:00:22Z)
• MaskDiff: Modeling Mask Distribution with Diffusion Probabilistic Model for Few-Shot Instance Segmentation [31.648523213206595]
  Few-shot instance segmentation extends the few-shot learning paradigm to the instance segmentation task. Conventional approaches have attempted to address the task via prototype learning, known as point estimation. We propose a novel approach, dubbed MaskDiff, which models the underlying conditional distribution of a binary mask.
  arXiv  Detail & Related papers  (2023-03-09T08:24:02Z)
• Exploiting Temporal Structures of Cyclostationary Signals for Data-Driven Single-Channel Source Separation [98.95383921866096]
  We study the problem of single-channel source separation (SCSS) We focus on cyclostationary signals, which are particularly suitable in a variety of application domains. We propose a deep learning approach using a U-Net architecture, which is competitive with the minimum MSE estimator.
  arXiv  Detail & Related papers  (2022-08-22T14:04:56Z)
• A research framework for writing differentiable PDE discretizations in JAX [3.4389358108344257]
  Differentiable simulators are an emerging concept with applications in several fields, from reinforcement learning to optimal control. We propose a library of differentiable operators and discretizations, by representing operators as mappings between families of continuous functions, parametrized by finite vectors. We demonstrate the approach on an acoustic optimization problem, where the Helmholtz equation is discretized using Fourier spectral methods, and differentiability is demonstrated using gradient descent to optimize the speed of sound of an acoustic lens.
  arXiv  Detail & Related papers  (2021-11-09T15:58:44Z)
• Pareto-wise Ranking Classifier for Multi-objective Evolutionary Neural Architecture Search [15.454709248397208]
  This study focuses on how to find feasible deep models under diverse design objectives. We propose a classification-wise Pareto evolution approach for one-shot NAS, where an online classifier is trained to predict the dominance relationship between the candidate and constructed reference architectures. We find a number of neural architectures with different model sizes ranging from 2M to 6M under diverse objectives and constraints.
  arXiv  Detail & Related papers  (2021-09-14T13:28:07Z)
• Multi-Source Domain Adaptation for Object Detection [52.87890831055648]
  We propose a unified Faster R-CNN based framework, termed Divide-and-Merge Spindle Network (DMSN) DMSN can simultaneously enhance domain innative and preserve discriminative power. We develop a novel pseudo learning algorithm to approximate optimal parameters of pseudo target subset.
  arXiv  Detail & Related papers  (2021-06-30T03:17:20Z)
• Distribution-Based Invariant Deep Networks for Learning Meta-Features [2.179313476241343]
  Recent advances in deep learning from probability distributions successfully achieve classification or regression from distribution samples, thus invariant under permutation of the samples. The proposed architecture, called Dida, inherits the NN properties of universal approximation, and its robustness w.r.t. Lipschitz-bounded transformations of the input distribution are established. The paper empirically and comparatively demonstrate the merits of the approach on two tasks defined at the dataset level.
  arXiv  Detail & Related papers  (2020-06-24T13:15:50Z)
• DC-NAS: Divide-and-Conquer Neural Architecture Search [108.57785531758076]
  We present a divide-and-conquer (DC) approach to effectively and efficiently search deep neural architectures. We achieve a $75.1%$ top-1 accuracy on the ImageNet dataset, which is higher than that of state-of-the-art methods using the same search space.
  arXiv  Detail & Related papers  (2020-05-29T09:02:16Z)

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.