Learning Permutation from Structure Without Supervision

• URL: http://arxiv.org/abs/2605.25551v1
• Date: Mon, 25 May 2026 08:08:47 GMT
• Title: Learning Permutation from Structure Without Supervision
• Abstract summary: Learning from structure without supervision induces a non-uniform uncertainty.<n>Existing methods control this process using a single global temperature.<n>We introduce an entropy-adaptive formulation of Gumbel-Sinkhorn that locally modulates temperature based on assignment uncertainty.
• Score: 16.16926085760151
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Many learning problems require uncovering a hidden ordering that reveals structure in unordered data, such as monotonicity in sorting or spatial continuity in jigsaw reconstruction. In these settings, permutations can be learned as latent operators by optimizing objectives defined directly on the reordered output, often without access to ground-truth orderings. Differentiable relaxations such as Gumbel-Sinkhorn make this approach practical by approximating permutation matrices with doubly stochastic matrices. However, learning from structure without supervision induces a non-uniform uncertainty: some assignments become confident early, while others remain ambiguous. Existing methods control this process using a single global temperature, forcing all assignments to sharpen or diffuse simultaneously and leading to instability at scale. We introduce an entropy-adaptive formulation of Gumbel-Sinkhorn that locally modulates temperature based on assignment uncertainty. This allows confident assignments to discretize early while preserving exploration where uncertainty remains. Across sorting and jigsaw reconstruction tasks and in routing-style settings, adaptive entropy control improves training stability and final permutation quality relative to fixed-temperature baselines, particularly as problem size and assignment ambiguity increase.

Related papers

• Stabilizing Extrapolation in Looped Transformers via Learned Stochastic Stopping [56.14767235650558]
  Looped Transformers, which repeatedly apply a shared transformer block, are an architecturally natural fit for variable-length computation tasks.<n>We trace this variance to the spurious correlation in simple algorithmic tasks between sequence length and number of loops.<n>Our work suggests that "when to stop" should be treated as a training design choice, not merely an inference-time-allocation rule.
  arXiv  Detail & Related papers  (2026-06-29T08:58:09Z)
• Dangerous Liaisons of Convex Learning and Non-Affine Aggregation [19.624245500772027]
  Last-iterate convergence and generalization guarantees in first-order convex learning hinge on the monotonicity of the update operator.<n>Non-affine aggregation prevents steady convergence and substantially degrades algorithmic stability.<n>Our results provide a unified theoretical framework explaining the disparate failure modes observed in modern learning systems.
  arXiv  Detail & Related papers  (2026-06-26T14:24:11Z)
• Variational Learning for Insertion-based Generation [50.15012125719375]
  We introduce a probabilistic framework for learning insertion order in variable-length insertion models.<n>We propose the Insertion Process (IP), a generative model that jointly learns where to insert, what to insert, and when to terminate.
  arXiv  Detail & Related papers  (2026-06-01T11:59:46Z)
• Learning Unbiased Permutations via Flow Matching [22.51828421737954]
  PermFlow is a conditional flow matching framework for learning permutations.<n>It captures multimodal permutation distributions rather than collapsing them to a single mode.<n>It achieves high accuracy on unambiguous inputs and recovers both valid permutations under ambiguity.
  arXiv  Detail & Related papers  (2026-05-16T02:10:35Z)
• Explicit Dropout: Deterministic Regularization for Transformer Architectures [55.09895958546215]
  Dropout is a widely used regularization technique in deep learning, but its effects are typically realized through masking.<n>We propose a deterministic formulation that expresses dropout as an additive regularizer directly incorporated into the training loss.
  arXiv  Detail & Related papers  (2026-04-22T12:45:51Z)
• Task Switching Without Forgetting via Proximal Decoupling [43.88767637201092]
  In continual learning, the primary challenge is to learn new information without forgetting old knowledge.<n>A common solution addresses this trade-off through regularization, penalizing changes to parameters critical for previous tasks.<n>We propose a different approach that separates task learning from stability enforcement via operator splitting.
  arXiv  Detail & Related papers  (2026-04-20T21:28:13Z)
• From Order to Distribution: A Spectral Characterization of Forgetting in Continual Learning [28.30104646642165]
  citetevron2022catastrophic analyzes forgetting under random orderings of a fixed task collection.<n>We study an exact-fit linear regime in which tasks are sampled i.i.d. from a task distribution$$, and ask how the generating distribution itself governs forgetting.
  arXiv  Detail & Related papers  (2026-04-15T04:29:00Z)
• Learn to Rank: Visual Attribution by Learning Importance Ranking [58.69028273772474]
  We propose a new approach to generate visual attribution maps for computer vision models.<n>Our method produces dense, pixel-level attributions in a single forward pass with optional, few-step gradient refinement.<n>Our experiments demonstrate consistent quantitative improvements and sharper, boundary-aligned explanations.
  arXiv  Detail & Related papers  (2026-04-07T12:53:22Z)
• Spectral Imbalance Causes Forgetting in Low-Rank Continual Adaptation [58.3773038915023]
  Continual learning aims to adapt pre-trained models to sequential tasks without forgetting previously acquired knowledge.<n>Most existing approaches treat continual learning as avoiding interference with past updates, rather than considering what properties make the current task-specific update naturally preserve previously acquired knowledge.<n>We address this problem using a projected first-order method compatible with standard deep-dots used in vision-language models.
  arXiv  Detail & Related papers  (2026-01-31T13:27:02Z)
• Generating Sequences by Learning to Self-Correct [64.0249217590888]
  Self-Correction decouples an imperfect base generator from a separate corrector that learns to iteratively correct imperfect generations. We show that Self-Correction improves upon the base generator in three diverse generation tasks.
  arXiv  Detail & Related papers  (2022-10-31T18:09:51Z)
• Leveraging Recursive Gumbel-Max Trick for Approximate Inference in Combinatorial Spaces [4.829821142951709]
  Structured latent variables allow incorporating meaningful prior knowledge into deep learning models. Standard learning approach is to define a latent variable as an algorithm output and to use a differentiable surrogate for training. We extend the Gumbel-Max trick to define distributions over structured domains.
  arXiv  Detail & Related papers  (2021-10-28T12:46:10Z)
• Discovering Non-monotonic Autoregressive Orderings with Variational Inference [67.27561153666211]
  We develop an unsupervised parallelizable learner that discovers high-quality generation orders purely from training data. We implement the encoder as a Transformer with non-causal attention that outputs permutations in one forward pass. Empirical results in language modeling tasks demonstrate that our method is context-aware and discovers orderings that are competitive with or even better than fixed orders.
  arXiv  Detail & Related papers  (2021-10-27T16:08:09Z)
• Self-Supervised Learning of Audio Representations from Permutations with Differentiable Ranking [35.644680012765356]
  We advance self-supervised learning from permutations, by pre-training a model to reorder shuffled parts of the spectrogram of an audio signal. We show that inverting permutations is a meaningful pretext task for learning audio representations in an unsupervised fashion.
  arXiv  Detail & Related papers  (2021-03-17T19:36:04Z)
• Provably End-to-end Label-Noise Learning without Anchor Points [118.97592870124937]
  We propose an end-to-end framework for solving label-noise learning without anchor points. Our proposed framework can identify the transition matrix if the clean class-posterior probabilities are sufficiently scattered.
  arXiv  Detail & Related papers  (2021-02-04T03:59:37Z)

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.