mlsanityData to 7 Oct 2026

Probabilistic Abduction for Visual Abstract Reasoning via Learning Rules in Vector-symbolic Architectures

arXiv 2401.16024 ↗First posted 29 Jan 2024cs.LG cs.AIcited by 9

On the map: LLM reasoning → Multimodal reasoning

Abstract (arXiv)

Abstract reasoning is a cornerstone of human intelligence, and replicating it with artificial intelligence (AI) presents an ongoing challenge. This study focuses on efficiently solving Raven's progressive matrices (RPM), a visual test for assessing abstract reasoning abilities, by using distributed computation and operators provided by vector-symbolic architectures (VSA). Instead of hard-coding the rule formulations associated with RPMs, our approach can learn the VSA rule formulations (hence the name Learn-VRF) with just one pass through the training data. Yet, our approach, with compact parameters, remains transparent and interpretable. Learn-VRF yields accurate predictions on I-RAVEN's in-distribution data, and exhibits strong out-of-distribution capabilities concerning unseen attribute-rule pairs, significantly outperforming pure connectionist baselines including large language models. Our code is available at https://github.com/IBM/learn-vector-symbolic-architectures-rule-formulations.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 25 of its 34 references that are in the map. First posted on arXiv 29 Jan 2024.

Builds on

25 of 34 references.

Closest ancestors · 1

Same direction, newest first

  1. 2022Deep Learning Methods for Abstract Visual Reasoning: A Survey on Raven's Progressive Matrices28 Jan

Other references · 24

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2019CLEVRER: CoLlision Events for Video REpresentation and Reasoningcited by 475
  3. 2019The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervisioncited by 369
  4. 2022Language Models Are Greedy Reasoners: A Systematic Formal Analysis of Chain-of-Thoughtcited by 334
  5. 2018Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understandingcited by 304
  6. 2022Emergent Analogical Reasoning in Large Language Modelscited by 239
  7. 2019RAVEN: A Dataset for Relational and Analogical Visual rEasoNingcited by 225
  8. 2021Abstraction and Analogy-Making in Artificial Intelligencecited by 103
  9. 2021A Survey on Hyperdimensional Computing aka Vector Symbolic Architectures, Part I: Models and Data Transformationscited by 98
  10. 2020Stratified Rule-Aware Network for Abstract Visual Reasoningcited by 72
  11. 2022A Neuro-vector-symbolic Architecture for Solving Raven's Progressive Matricescited by 64
  12. 2022Evaluating the Impact of Model Scale for Compositional Generalization in Semantic Parsingcited by 50
12 more
  1. 2020The Scattering Compositional Learner: Discovering Objects, Attributes, Relationships in Analogical Reasoningcited by 46
  2. 2020Scale-Localized Abstract Reasoningcited by 45
  3. 2022Neurocompositional computing: From the Central Paradox of Cognition to a new generation of AI systemscited by 44
  4. 2021Abstract Spatial-Temporal Reasoning via Probabilistic Abduction and Executioncited by 43
  5. 2020Visual Concept-Metaconcept Learningcited by 39
  6. 2019Enhancing the Transformer with Explicit Relational Encoding for Math Problem Solvingcited by 34
  7. 2018Learning to Reason with Third-Order Tensor Productscited by 27
  8. 2020Variable Binding for Sparse Distributed Representations: Theory and Applicationscited by 27
  9. 2021Techniques for Symbol Grounding with SATNetcited by 24
  10. 2023In-Context Analogical Reasoning with Pre-Trained Language Modelscited by 22
  11. 2022FALCON: Fast Visual Concept Learning by Integrating Images, Linguistic descriptions, and Conceptual Relationscited by 20
  12. 2022Knowledge-based Analogical Reasoning in Neuro-symbolic Latent Spacescited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 3; 2026: 1.

9 papers.

  1. 2024Limits of Transformer Language Models on Learning to Compose Algorithmscited by 13
  2. 2024Towards Learning Abductive Reasoning using VSA Distributed Representationscited by 9
  3. 2024Towards Learning to Reason: Comparing LLMs with Neuro-Symbolic on Arithmetic Relations in Abstract Reasoningcited by 7
  4. 2024Generalized Holographic Reduced Representationscited by 6
  5. 2025A Study on Neuro-Symbolic Artificial Intelligence: Healthcare Perspectivescited by 5
  6. 2024H3DFact: Heterogeneous 3D Integrated CIM for Factorization with Holographic Perceptual Representationscited by 2
  7. 2025Systematic Abductive Reasoning via Diverse Relation Representations in Vector-symbolic Architecturecited by 1
  8. 2025NSFlow: An End-to-End FPGA Framework with Scalable Dataflow Architecture for Neuro-Symbolic AIcited by 1
  9. 2026Gram-Space: Structure-Preserving Codebook Compression for Memory-Efficient Neuro-Symbolic AIcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Towards Generative Abstract Reasoning: Completing Raven's Progressive Matrix via Rule Abstraction and Selection
  2. 2023Learning Abstract Visual Reasoning via Task Decomposition: A Case Study in Raven Progressive Matrices
  3. 2025A Study of Rule Omission in Raven's Progressive Matrices
  4. 2023Learning Differentiable Logic Programs for Abstract Visual Reasoning
  5. 2024A-I-RAVEN and I-RAVEN-Mesh: Two New Benchmarks for Abstract Visual Reasoning
  6. 2023Abstracting Concept-Changing Rules for Solving Raven's Progressive Matrix Problems
  7. 2024A Feature-based Generalizable Prediction Model for Both Perceptual and Abstract Reasoning
  8. 2024Neural networks for abstraction and reasoning: Towards broad generalization in machines