mlsanityData to 14 Oct 2026

Neural Algorithmic Reasoning

arXiv 2105.02761 ↗First posted 6 May 2021cs.LG cs.AI cs.DScited by 94

Not on the map. Why?

Abstract (arXiv)

Algorithms have been fundamental to recent global technological advances and, in particular, they have been the cornerstone of technical advances in one field rapidly being applied to another. We argue that algorithms possess fundamentally different qualities to deep learning methods, and this strongly suggests that, were deep learning methods better able to mimic algorithms, generalisation of the sort seen with algorithms would become possible with deep learning -- something far out of the reach of current machine learning methods. Furthermore, by representing elements in a continuous space of learnt algorithms, neural networks are able to adapt known algorithms more closely to real-world problems, potentially finding more efficient and pragmatic solutions than those proposed by human computer scientists. Here we present neural algorithmic reasoning -- the art of building neural networks that are able to execute algorithmic computation -- and provide our opinion on its transformative potential for running classical algorithms on inputs previously considered inaccessible to them.

Key facts. Cited by 94 papers in the map (data through 14 Oct 2026). Builds on 9 of its 15 references that are in the map. First posted on arXiv 6 May 2021.

Builds on

9 of 15 references.

References · 9

  1. 2014Neural Turing Machinescited by 644
  2. 2021Combinatorial optimization and reasoning with graph neural networkscited by 268
  3. 2015Neural GPUs Learn Algorithmscited by 121
  4. 2019Neural Execution of Graph Algorithmscited by 121
  5. 2015Neural Programmer-Interpreterscited by 101
  6. 2020Pointer Graph Networkscited by 53
  7. 2015Neural Random-Access Machinescited by 34
  8. 2020XLVIN: eXecuted Latent Value Iteration Netscited by 17
  9. 2020Strong Generalization and Efficiency in Neural Programscited by 12

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 6; 2022: 14; 2023: 19; 2024: 24; 2025: 20; 2026: 11.

94 papers.

Sort:

By area:

  1. 2021Show Your Work: Scratchpads for Intermediate Computation with Language Modelscited by 886
  2. 2023Faith and Fate: Limits of Transformers on Compositionalitycited by 503
  3. 2023Large Language Models as Tool Makerscited by 303
  4. 2023What Algorithms can Transformers Learn? A Study in Length Generalizationcited by 199
  5. 2022Teaching Algorithmic Reasoning via In-context Learningcited by 112
  6. 2022The CLRS Algorithmic Reasoning Benchmarkcited by 106
  7. 2024Understanding Transformer Reasoning Capabilities via Graph Algorithmscited by 67
  8. 2022A Generalist Neural Algorithmic Learnercited by 59
  9. 2022Pareto Set Learning for Neural Multi-objective Combinatorial Optimizationcited by 56
  10. 2022Graph Neural Networks are Dynamic Programmerscited by 55
  11. 2024GraphArena: Evaluating and Exploring Large Language Models on Graph Computationcited by 49
  12. 2022Cell Attention Networkscited by 35
38 more
  1. 2023Neural Algorithmic Reasoning with Causal Regularisationcited by 31
  2. 2022Towards Better Out-of-Distribution Generalization of Neural Algorithmic Reasoning Taskscited by 30
  3. 2024Simulation of Graph Algorithms with Looped Transformerscited by 30
  4. 2021Neural Algorithmic Reasoners are Implicit Plannerscited by 21
  5. 2023Dual Algorithmic Reasoningcited by 21
  6. 2025Pareto Set Learning for Multi-Objective Reinforcement Learningcited by 20
  7. 2022Systematic Generalization and Emergent Structures in Transformers Trained on Structured Taskscited by 19
  8. 2023From Latent Graph to Latent Topology Inference: Differentiable Cell Complex Modulecited by 19
  9. 2023Neural Algorithmic Reasoning for Combinatorial Optimisationcited by 19
  10. 2024Limits of Deep Learning: Sequence Modeling through the Lens of Complexity Theorycited by 19
  11. 2023Neural Algorithmic Reasoning Without Intermediate Supervisioncited by 17
  12. 2024SymbolicAI: A framework for logic-based approaches combining generative models and solverscited by 16
  13. 2022Neural Set Function Extensions: Learning with Discrete Functions in High Dimensionscited by 14
  14. 2023CIN++: Enhancing Topological Message Passingcited by 14
  15. 2021Reasoning-Modulated Representationscited by 13
  16. 2023Asynchronous Algorithmic Alignment with Cocyclescited by 13
  17. 2024PUZZLES: A Benchmark for Neural Algorithmic Reasoningcited by 13
  18. 2023SALSA-CLRS: A Sparse and Scalable Benchmark for Algorithmic Reasoningcited by 12
  19. 2023Are Graph Neural Networks Optimal Approximation Algorithms?cited by 11
  20. 2024On the Markov Property of Neural Algorithmic Reasoning: Analyses and Methodscited by 11
  21. 2025Graph neural networks extrapolate out-of-distribution for shortest pathscited by 11
  22. 2023Recursive Algorithmic Reasoningcited by 9
  23. 2024Discrete Neural Algorithmic Reasoningcited by 9
  24. 2024Converging Paradigms: The Synergy of Symbolic and Connectionist AI in LLM-Empowered Autonomous Agentscited by 9
  25. 2025Primal-Dual Neural Algorithmic Reasoningcited by 9
  26. 2023Neural Priority Queues for Graph Neural Networkscited by 7
  27. 2022Learning Graph Algorithms With Recurrent Graph Neural Networkscited by 6
  28. 2023Parallel Algorithms Align with Neural Executioncited by 6
  29. 2024Recurrent Aggregators in Neural Algorithmic Reasoningcited by 6
  30. 2024Challenges of Generating Structurally Diverse Graphscited by 6
  31. 2024Measuring Diversity: Axioms and Challengescited by 6
  32. 2024Deep Equilibrium Algorithmic Reasoningcited by 6
  33. 2022Learning to Configure Computer Networks with Neural Algorithmic Reasoningcited by 5
  34. 2024Are Large-Language Models Graph Algorithmic Reasoners?cited by 5
  35. 2025Machine Learning meets Algebraic Combinatorics: A Suite of Datasets Capturing Research-level Conjecturing Ability in Pure Mathematicscited by 5
  36. 2022Knowledge-based Analogical Reasoning in Neuro-symbolic Latent Spacescited by 4
  37. 2022Continuous Neural Algorithmic Plannerscited by 4
  38. 2024Neural Algorithmic Reasoning with Multiple Correct Solutionscited by 4

50 most cited of 94 All 94 (JSON)