mlsanityData to 14 Oct 2026

Making Neural Programming Architectures Generalize via Recursion

arXiv 1704.06611 ↗First posted 21 Apr 2017cs.LG cs.NE cs.PLcited by 35

Not on the map. Why?

Abstract (arXiv)

Empirically, neural networks that attempt to learn programs from data have exhibited poor generalizability. Moreover, it has traditionally been difficult to reason about the behavior of these models beyond a certain level of input complexity. In order to address these issues, we propose augmenting neural architectures with a key abstraction: recursion. As an application, we implement recursion in the Neural Programmer-Interpreter framework on four tasks: grade-school addition, bubble sort, topological sort, and quicksort. We demonstrate superior generalizability and interpretability with small amounts of training data. Recursion divides the problem into smaller pieces and drastically reduces the domain of each neural network component, making it tractable to prove guarantees about the overall system's behavior. Our experience suggests that in order for neural architectures to robustly learn program semantics, it is necessary to incorporate a concept like recursion.

Key facts. Cited by 35 papers in the map (data through 14 Oct 2026). First posted on arXiv 21 Apr 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2018: 1; 2019: 0; 2020: 0; 2021: 11; 2022: 7; 2023: 10; 2024: 3; 2025: 1; 2026: 2.

35 papers.

Sort:

By area:

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2021Measuring Coding Challenge Competence With APPScited by 902
  3. 2022Solving math word problems with process- and outcome-based feedbackcited by 901
  4. 2023WizardMath: Empowering Mathematical Reasoning for Large Language Models via Reinforced Evol-Instructcited by 625
  5. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 558
  6. 2018Scalable agent alignment via reward modeling: a research directioncited by 447
  7. 2023Teaching Arithmetic to Small Transformerscited by 123
  8. 2024Beyond A*: Better Planning with Transformers via Search Dynamics Bootstrappingcited by 90
  9. 2021Neural Production Systems: Learning Rule-Governed Visual Dynamicscited by 75
  10. 2023Large Language Models can Learn Rulescited by 69
  11. 2021Learning to Synthesize Programs as Interpretable and Generalizable Policiescited by 59
  12. 2023Learning Transformer Programscited by 41
23 more
  1. 2024Learning Iterative Reasoning through Energy Diffusioncited by 36
  2. 2022Towards Data-and Knowledge-Driven Artificial Intelligence: A Survey on Neuro-Symbolic Computingcited by 35
  3. 2022Learning Iterative Reasoning through Energy Minimizationcited by 32
  4. 2021Pointer Value Retrieval: A new benchmark for understanding the limits of neural network generalizationcited by 29
  5. 2022How to Reuse and Compose Knowledge for a Lifetime of Tasks: A Survey on Continual Learning and Functional Compositioncited by 27
  6. 2023Recursion of Thought: A Divide-and-Conquer Approach to Multi-Context Reasoning with Language Modelscited by 24
  7. 2021Neural-guided, Bidirectional Program Search for Abstraction and Reasoningcited by 23
  8. 2021Pushing the Limits of Rule Reasoning in Transformers through Natural Language Satisfiabilitycited by 18
  9. 2023Recursive Algorithmic Reasoningcited by 9
  10. 2021Recognizing and Verifying Mathematical Equations using Multiplicative Differential Neural Unitscited by 8
  11. 2023Hierarchical Neural Program Synthesiscited by 7
  12. 2021Neural Sequence-to-grid Module for Learning Symbolic Rulescited by 5
  13. 2021Evolutionary Training and Abstraction Yields Algorithmic Generalization of Neural Computerscited by 5
  14. 2026Gradient-Based Program Synthesis with Neurally Interpreted Languagescited by 5
  15. 2022Synthesizing Programs with Continuous Optimizationcited by 2
  16. 2021State-Space Constraints Can Improve the Generalisation of the Differentiable Neural Computer to Input Sequences With Unseen Lengthcited by 1
  17. 2023Neuro-Symbolic Sudoku Solvercited by 1
  18. 2026Training Transformers as a Universal Computercited by 1
  19. 2021Understanding in Artificial Intelligencecited by 0
  20. 2023Learning-Based Automatic Synthesis of Software Code and Configurationcited by 0
  21. 2023OC-NMN: Object-centric Compositional Neural Module Network for Generative Visual Analogical Reasoningcited by 0
  22. 2024Bridging the Gap: Representation Spaces in Neuro-Symbolic AIcited by 0
  23. 2025Learning Model Successorscited by 0