mlsanityData to 7 Oct 2026

Unlocking Slot Attention by Changing Optimal Transport Costs

arXiv 2301.13197 ↗First posted 30 Jan 2023cs.LG cs.CVcited by 16

Not on the map. Why?

Abstract (arXiv)

Slot attention is a powerful method for object-centric modeling in images and videos. However, its set-equivariance limits its ability to handle videos with a dynamic number of objects because it cannot break ties. To overcome this limitation, we first establish a connection between slot attention and optimal transport. Based on this new perspective we propose MESH (Minimize Entropy of Sinkhorn): a cross-attention module that combines the tiebreaking properties of unregularized optimal transport with the speed of regularized optimal transport. We evaluate slot attention using MESH on multiple object-centric learning benchmarks and find significant improvements over slot attention in every setting.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Builds on 33 of its 50 references that are in the map. First posted on arXiv 30 Jan 2023.

Builds on

33 of 50 references.

References · 33

  1. 2017Attention Is All You Needcited by 59,932
  2. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,277
  3. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
  4. 2016CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoningcited by 1,445
  5. 2017Deep Setscited by 1,394
  6. 2014On the Properties of Neural Machine Translation: Encoder-Decoder Approachescited by 1,303
  7. 2020Object-Centric Learning with Slot Attentioncited by 865
  8. 2018Computational Optimal Transportcited by 765
  9. 2019Deep Equilibrium Modelscited by 603
  10. 2019CLEVRER: CoLlision Events for Video REpresentation and Reasoningcited by 475
  11. 2019Multi-Object Representation Learning with Iterative Variational Inferencecited by 359
  12. 2021Conditional Object-Centric Learning from Videocited by 247
21 more
  1. 2019Differentiation of Blackbox Combinatorial Solverscited by 224
  2. 2019Recurrent Independent Mechanismscited by 219
  3. 2020SPACE: Unsupervised Object-Oriented Scene Representation via Spatial Attention and Decompositioncited by 190
  4. 2022SlotFormer: Unsupervised Visual Dynamics Simulation with Object-Centric Modelscited by 146
  5. 2021Sinkformers: Transformers with Doubly Stochastic Attentioncited by 136
  6. 2016On Differentiating Parameterized Argmin and Argmax Problems with Application to Bi-level Optimizationcited by 108
  7. 2021ClevrTex: A Texture-Rich Benchmark for Unsupervised Multi-Object Segmentationcited by 93
  8. 2022Object Scene Representation Transformercited by 90
  9. 2021JFB: Jacobian-Free Backpropagation for Implicit Networkscited by 84
  10. 2017Smooth and Sparse Optimal Transportcited by 83
  11. 2019Deep Set Prediction Networkscited by 71
  12. 2022Re-basin via implicit Sinkhorn differentiationcited by 61
  13. 2020Generative Neurosymbolic Machinescited by 57
  14. 2021Systematic Evaluation of Causal Discovery in Visual Model Based Reinforcement Learningcited by 53
  15. 2021Unsupervised Layered Image Decomposition into Object Prototypescited by 50
  16. 2020Object Files and Schemata: Factorizing Declarative and Procedural Knowledge in Dynamical Systemscited by 48
  17. 2022Object Representations as Fixed Points: Training Iterative Refinement Algorithms with Implicit Differentiationcited by 46
  18. 2021MarioNette: Self-Supervised Sprite Learningcited by 35
  19. 2020SCOUTER: Slot Attention-based Classifier for Explainable Image Recognitioncited by 29
  20. 2021Multiset-Equivariant Set Prediction with Approximate Implicit Differentiationcited by 17
  21. 2022Differentiable Mathematical Programming for Object-Centric Representation Learningcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 3; 2025: 6; 2026: 4.

16 papers.

Sort:

By area:

  1. 2023Object-centric architectures enable efficient causal representation learningcited by 21
  2. 2023Object-Centric Learning with Slot Mixture Modulecited by 11
  3. 2024Unsupervised Concept Discovery Mitigates Spurious Correlationscited by 11
  4. 2025PEEK: Guiding and Minimal Image Representations for Zero-Shot Generalization of Robot Manipulation Policiescited by 11
  5. 2024On permutation-invariant neural networkscited by 10
  6. 2025ESPFormer: Doubly-Stochastic Attention with Expected Sliced Transport Planscited by 10
  7. 2023Reusable Slotwise Mechanismscited by 4
  8. 2025Neurosymbolic Object-Centric Learning with Distant Supervisioncited by 2
  9. 2024Attention Normalization Impacts Cardinality Generalization in Slot Attentioncited by 1
  10. 2025SMARTe: Slot-based Method for Accountable Relational Triple extractioncited by 1
  11. 2025Hierarchical Compact Clustering Attention (COCA) for Unsupervised Object-Centric Learningcited by 1
  12. 2026Mitigating Simplicity Bias in OOD Detection through Object Co-occurrence Analysiscited by 1
4 more
  1. 2025Unsupervised Evolutionary Cell Type Matching via Entropy-Minimized Optimal Transportcited by 0
  2. 2026Improving Set Function Approximation with Quasi-Arithmetic Neural Networkscited by 0
  3. 2026When Attention Collapses: Residual Evidence Modeling for Compositional Inferencecited by 0
  4. 2026Spacetime Optimal-Transport Attention for Visuo-Haptic Imitation Learning of Contact-Rich Manipulationcited by 0