CORDS: Continuous Representations of Discrete Structures

• URL: http://arxiv.org/abs/2601.21583v1
• Date: Thu, 29 Jan 2026 11:46:17 GMT
• Title: CORDS: Continuous Representations of Discrete Structures
• Authors: Tin Hadži Veljković, Erik Bekkers, Michael Tiemann, Jan-Willem van de Meent,
• Abstract summary: We present a novel strategy for casting prediction of variable-sized sets as a continuous inference problem.<n>Our approach, CORDS, provides an invertible mapping that transforms a set of spatial objects into continuous fields.<n>Because the mapping is invertible, models operate entirely in field space while remaining exactly decodable to discrete sets.
• Score: 7.525463742653151
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Many learning problems require predicting sets of objects when the number of objects is not known beforehand. Examples include object detection, molecular modeling, and scientific inference tasks such as astrophysical source detection. Existing methods often rely on padded representations or must explicitly infer the set size, which often poses challenges. We present a novel strategy for addressing this challenge by casting prediction of variable-sized sets as a continuous inference problem. Our approach, CORDS (Continuous Representations of Discrete Structures), provides an invertible mapping that transforms a set of spatial objects into continuous fields: a density field that encodes object locations and count, and a feature field that carries their attributes over the same support. Because the mapping is invertible, models operate entirely in field space while remaining exactly decodable to discrete sets. We evaluate CORDS across molecular generation and regression, object detection, simulation-based inference, and a mathematical task involving recovery of local maxima, demonstrating robust handling of unknown set sizes with competitive accuracy.

Related papers

• Spatial-Temporal-Spectral Unified Modeling for Remote Sensing Dense Prediction [20.1863553357121]
  Current deep learning architectures for remote sensing are fundamentally rigid.<n>We introduce the Spatial-Temporal-Spectral Unified Network (STSUN) for unified modeling.<n> STSUN can adapt to input and output data with arbitrary spatial sizes, temporal lengths, and spectral bands.<n>It unifies various dense prediction tasks and diverse semantic class predictions.
  arXiv  Detail & Related papers  (2025-05-18T07:39:17Z)
• Open-Set Semantic Uncertainty Aware Metric-Semantic Graph Matching [10.439907158831303]
  A metric of semantic uncertainty for open-set object detections is calculated and incorporated into an object-level uncertainty tracking framework. The proposed methods are feasible for real-time use in marine environments for the robust, open-set, multi-object, semantic-uncertainty-aware loop closure detection.
  arXiv  Detail & Related papers  (2024-09-17T20:53:47Z)
• Object-centric architectures enable efficient causal representation learning [51.6196391784561]
  We show that when the observations are of multiple objects, the generative function is no longer injective and disentanglement fails in practice. We develop an object-centric architecture that leverages weak supervision from sparse perturbations to disentangle each object's properties. This approach is more data-efficient in the sense that it requires significantly fewer perturbations than a comparable approach that encodes to a Euclidean space.
  arXiv  Detail & Related papers  (2023-10-29T16:01:03Z)
• Learning Invariant Molecular Representation in Latent Discrete Space [52.13724532622099]
  We propose a new framework for learning molecular representations that exhibit invariance and robustness against distribution shifts. Our model achieves stronger generalization against state-of-the-art baselines in the presence of various distribution shifts.
  arXiv  Detail & Related papers  (2023-10-22T04:06:44Z)
• Vector Quantisation for Robust Segmentation [14.477470283239501]
  The reliability of segmentation models in the medical domain depends on the model's robustness to perturbations in the input space. We propose and justify that learning a discrete representation in a low dimensional embedding space improves robustness of a segmentation model. This is achieved with a dictionary learning method called vector quantisation.
  arXiv  Detail & Related papers  (2022-07-05T09:52:53Z)
• A Bayesian Treatment of Real-to-Sim for Deformable Object Manipulation [59.29922697476789]
  We propose a novel methodology for extracting state information from image sequences via a technique to represent the state of a deformable object as a distribution embedding. Our experiments confirm that we can estimate posterior distributions of physical properties, such as elasticity, friction and scale of highly deformable objects, such as cloth and ropes.
  arXiv  Detail & Related papers  (2021-12-09T17:50:54Z)
• SPANet: Generalized Permutationless Set Assignment for Particle Physics using Symmetry Preserving Attention [62.43586180025247]
  Collisions at the Large Hadron Collider produce variable-size sets of observed particles. Physical symmetries of decay products complicate assignment of observed particles to decay products. We introduce a novel method for constructing symmetry-preserving attention networks.
  arXiv  Detail & Related papers  (2021-06-07T18:18:20Z)
• SIMstack: A Generative Shape and Instance Model for Unordered Object Stacks [38.042876641457255]
  We propose a depth-conditioned Variational Auto-Encoder (VAE) trained on a dataset of objects stacked under physics simulation. We formulate instance segmentation as a centre voting task which allows for class-agnostic detection and doesn't require setting the maximum number of objects in the scene. Our method has practical applications in providing robots some of the ability humans have to make rapid intuitive inferences of partially observed scenes.
  arXiv  Detail & Related papers  (2021-03-30T15:42:43Z)
• Evidential Sparsification of Multimodal Latent Spaces in Conditional Variational Autoencoders [63.46738617561255]
  We consider the problem of sparsifying the discrete latent space of a trained conditional variational autoencoder. We use evidential theory to identify the latent classes that receive direct evidence from a particular input condition and filter out those that do not. Experiments on diverse tasks, such as image generation and human behavior prediction, demonstrate the effectiveness of our proposed technique.
  arXiv  Detail & Related papers  (2020-10-19T01:27:21Z)
• Spatial Classification With Limited Observations Based On Physics-Aware Structural Constraint [18.070762916388272]
  spatial classification with limited feature observations has been a challenging problem in machine learning. This paper extends our recent approach by allowing feature values of samples in each class to follow a multi-modal distribution. We propose learning algorithms for the extended model with multi-modal distribution.
  arXiv  Detail & Related papers  (2020-08-25T20:07:28Z)
• Closed-Form Factorization of Latent Semantics in GANs [65.42778970898534]
  A rich set of interpretable dimensions has been shown to emerge in the latent space of the Generative Adversarial Networks (GANs) trained for synthesizing images. In this work, we examine the internal representation learned by GANs to reveal the underlying variation factors in an unsupervised manner. We propose a closed-form factorization algorithm for latent semantic discovery by directly decomposing the pre-trained weights.
  arXiv  Detail & Related papers  (2020-07-13T18:05:36Z)

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.