Pith. sign in

REVIEW 3 major objections 6 minor 43 references

Systematic Abductive Reasoning via Diverse Relation Representations in Vector-symbolic Architecture

T0 review · 3 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Rel-SAR shows that a vector-symbolic architecture using numeric, circular, and boolean high-dimensional codes can solve Raven's Progressive Matrices at 96.5% average accuracy while keeping abduction and execution interpretable.

desk verdict A genuinely better VSA-based RPM solver with a real win on position rules, but the 'systematic abduction' claim rests on supervised rule labels and an unverified assumption about RAVEN's rule vocabulary. read the letter →

arxiv 2501.11896 v2 pith:JQQADV7J submitted 2025-01-21 cs.AI

Zhong-Hua Sun , Ru-Yuan Zhang , Zonglei Zhen , Da-Hui Wang , Yong-Jie Li , Xiaohong Wan , Hongzhi You This is my paper · ORCID
classification cs.AI
keywords abstractvisualreasoningRaven'sProgressiveMatricesvector-symbolicarchitecturehyperdimensionalcomputingrelationrepresentationruleabductionneuro-symbolicout-of-distributiongeneralization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that abstract visual reasoning in Raven's Progressive Matrices can be carried out by a vector-symbolic architecture in which every attribute value is represented as a high-dimensional vector with explicit numeric, periodic, or logical semantics, and rules are abducted and executed by two general relation functions. The proposed model, Rel-SAR, reaches 96.5% average accuracy on RAVEN and 98.0% on I-RAVEN without attribute labels, on par with the best monolithic deep networks, while improving over the prior neuro-symbolic model NVSA by 8.8 and 9.9 points on average. The largest gains come on grid configurations where rules act on object positions, up 39.1 points on 3x3Grid and 20.8 points on Out-InGrid. If the approach is right, interpretable symbolic reasoning and high perceptual accuracy do not have to be traded off, and the same algebraically grounded vector code can support perception, rule finding, and rule execution in one mechanism.

What carries the argument

The load-bearing machinery is the pairing of four atomic vector types with two relation functions. Numeric vectors, built by fractional power encoding, make real addition correspond to vector binding; circular vectors add periodicity, which is exactly what cyclic grid shifts need; boolean vectors implement NOT, AND, OR, and (in the full version) XOR through binding and similarity operations. The structured high-dimensional representation (SHDR) is a role-filler binding of position vectors and attribute vectors that lets a whole image panel be encoded in one vector and then decomposed by unbinding. The numerical relation function $r_{\mathrm{Num}} = \bigcirc_{i=1}^{N} v_i^{\circ \mathrm{op}_i}$ and the logical relation function $r_{\mathrm{Lgc}} = (\mathrm{op}_1 v_1 \wedge \mathrm{op}_2 v_2) \circ \mathrm{op}_3 v_3$ take several attribute vectors and operator powers as input and return an output vector; rule abduction searches for one parameter setting that makes all relation outputs agree, while rule execution uses the inverse functions to predict the missing panel. A 5-operator full logical version in Appendix A covers XOR, though RAVEN itself only needs the simplified form for its Arithmetic+ and Arithmetic- position rules.

What would settle it

Construct an RPM-style test where object existence follows the XOR rule ($v_3 = v_1 \oplus v_2$) and train Rel-SAR end-to-end on it. The paper's own Appendix A states that the simplified logical relation function cannot represent XOR, so if the model fails on this rule, the claimed systematicity over logical rules is disproved.

Watch

Extended reading notes

Core claim

Rel-SAR is a neuro-symbolic system that learns a structured high-dimensional representation (SHDR) of each image panel through VSA binding and bundling: a ResNet-50 predicts the SHDR, which is then decomposed into per-object attribute vectors by unbinding. A query-attention step re-embeds the noisy neural estimates into backend codebooks whose vectors carry algebraic semantics: numeric vectors for type, size, color, and number; circular vectors for grid positions; boolean vectors for object existence. Two relation functions are the rule representation: the numerical function $R_{\mathrm{Num}}(v_{1:N},\mathrm{OP}_{1:M}) = \bigcirc_{i=1}^{N} v_i^{\circ \mathrm{op}_i}$ and the simplified logical function $R_{\mathrm{Lgc}}(v_{1:N},\mathrm{OP}_{1:M}) = (\mathrm{op}_1 v_1 \wedge \mathrm{op}_2 v_2) \circ \mathrm{op}_3 v_3$. Rule abduction finds operator powers $\mathrm{OP}_{1:M}$ that make all pairwise row and column relation outputs agree, and rule execution inverts the same functions to predict the missing panel. The paper reports that this unified scheme reaches 96.5% on RAVEN and 98.0% on I-RAVEN, and that with ground-truth attribute labels its reasoning backend scores 99.2% on I-RAVEN, 6.8 points above ARLC. The central claim is that systematicity comes from diverse relation representations rather than from a lookup table of per-rule templates.

Load-bearing premise

The model only works if every rule in the puzzle can be expressed by adding together vectors of a few fixed types; the authors acknowledge in Appendix A that the simplified logical function cannot express XOR, so any rule of that form lies outside the model's reach.

Editorial extensions

If this is right

  • Because the same relation functions are used across attributes, a rule learned for one attribute can transfer to another: the paper reports near-perfect out-of-distribution transfer for Type and Size rules on I-RAVEN.
  • Position-dependent configurations stop being the weak spot of vector-symbolic solvers: 3x3Grid rises from 57.1% to 96.2% and Out-InGrid from 65.4% to 86.2% on RAVEN compared with NVSA.
  • End-to-end training without attribute labels is sufficient: Rel-SAR's accuracy stays nearly identical when attribute labels are removed, whereas NVSA and ALANS degrade substantially.
  • With ground-truth attribute labels, the reasoning backend reaches 99.2% on I-RAVEN, surpassing prior VSA-based rule-learning models ARLC and Learn-VRF.
  • The abducted rules are explicit operator-power and output-vector pairs, so the final answer choice can be inspected as a series of vector operations.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural extension beyond RPM is to other relational reasoning benchmarks such as Bongard problems or visual analogy tasks, where the full logical relation function from Appendix A would be needed to handle XOR-style rules that RAVEN does not contain.
  • Because the numerical relation function only expresses additive relations, a testable extension is to represent attribute values in a logarithmic FPE basis so that vector binding encodes multiplication and ratios, which would let the same architecture handle proportion-style rules.
  • The reported limits on Color out-of-distribution transfer could be addressed by normalizing attribute-value ranges before FPE encoding, so that a rule learned on a narrow range like Type transfers to a wider range like Color.
  • The query-attention bridge between task-independent frontend codebooks and algebraically meaningful backend codebooks suggests a general recipe: train perception against random codebooks, then re-express the result in a structured codebook, a pattern that could apply to other neuro-symbolic systems.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper proposes Rel-SAR, a neuro-symbolic model for Raven's Progressive Matrices (RPM) built on vector-symbolic architectures (VSA). It introduces four types of atomic high-dimensional vectors (random, numeric, circular, boolean), a structured high-dimensional representation (SHDR) for image panels and grid components, and numerical and logical relation functions that support rule abduction and execution. The model uses a ResNet-50 perception frontend to predict panel SHDRs and a reasoning backend that abducts rule parameters (operator powers OP and output r) via relation functions, executes them with inverse relation functions, and selects the answer with highest similarity. Experiments report 96.5% average accuracy on RAVEN and 98.0% on I-RAVEN under end-to-end training, as well as 99.2% on I-RAVEN when using attribute labels in a reasoning-only evaluation. The paper also reports attribute-value and OOD generalization experiments.

Significance. If the results hold, Rel-SAR is a meaningful step toward interpretable and systematic abstract visual reasoning: it substantially improves over prior VSA-based neuro-symbolic models on position-dependent configurations, and its perception frontend achieves near-perfect attribute extraction. The algebraic formulation of relation functions (Eqs. 6-9 and the full logical version in Appendix A) is internally consistent, and the experimental comparisons are broad. However, the central claim of systematic abductive reasoning is qualified by two issues: the main end-to-end results are obtained with ground-truth rule labels, and the simplified logical relation function cannot represent XOR, with no evidence that RAVEN lacks such rules. The reported OOD generalization is also less robust than the abstract suggests.

major comments (3)
  1. [IV-F.1, Eqs. (23)-(24), Table VI] The main end-to-end comparison in Table VI uses rule labels through the losses Lop (Eq. 23) and Lr (Eq. 24), where the rule learners regress directly to ground-truth operator powers OP_gt and relation outputs r_gt. This supervision is disclosed in Section IV-F.1 but is not flagged in the abstract or in Section V-B, where the results are described simply as 'end-to-end learning.' Because the rule abduction module is therefore not autonomously deriving rules from the context panels alone, the high accuracy in Table VI largely reflects this extra supervision. Please report a variant trained without rule labels (for example, using only answer labels and the Lrs objective of Eq. 27) and clearly state in the abstract and results section what supervision each reported accuracy uses.
  2. [Appendix A, Eq. (7), Table IV] The simplified logical relation function RLgc (Eq. 7) cannot express XOR, and the paper asserts that 'the RAVEN dataset only involve Arithmetic+ and Arithmetic- logical rules' without providing per-rule statistics or dataset annotations. This assumption is load-bearing because the reported accuracy on position-dependent configurations (3x3Grid, Out-InGrid) would drop if any Position rule were XOR, as the reasoning backend could not represent or abduct it. The paper already defines the full logical relation function that can express XOR (Eq. 29) but never uses it in experiments. Please either provide dataset rule statistics demonstrating that XOR and other non-simplified rules are absent, or run the experiments with the full logical relation function.
  3. [Abstract; Section V-D, Table XI] The abstract claims 'robust out-of-distribution generalization,' but Table XI shows that Rel-SAR's OOD accuracy is markedly below that of Learn-VRF and ARLC on the Color attribute (e.g., Color Distribute Three: 70.7% vs. 100% and 99.8%). The paper itself acknowledges this limitation in the conclusion, but the abstract's unconditional claim is too strong. Please either temper the abstract's claim or add analysis explaining why Color fails and whether this is inherent to the relation-function representation or a property of the rule learner.
minor comments (6)
  1. [Abstract] There is a typo: 'high-dimentional' should be 'high-dimensional.'
  2. [Conclusion] The word 'sytematic' in the final section should be 'systematic.'
  3. [Throughout] The configuration name is inconsistently written as 'Out-InGrid' in the text and 'O-InGrid' in several table headers, figures, and the caption of Table X; please use one notation consistently.
  4. [IV-B.1, Eq. (6)] The notation v(◦opi)i is easy to misread as exponentiation; consider adding a sentence clarifying that the superscript denotes repeated binding (op_i times).
  5. [Table II] The OR operation is defined using AND and XOR; a short derivation or reference would help readers verify that this implements Boolean OR in the vector domain.
  6. [V] No code or data are provided, which limits reproducibility of the reported numbers; the authors should consider releasing the implementation and dataset rule statistics.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found; the central claim is independently supported by held-out RAVEN/I-RAVEN benchmarks and a consistency-only reasoning experiment.

full rationale

The derivation chain is not circular. The relation functions (Eqs. 6-9) are hand-specified VSA algebraic operations, and the inverse functions are their exact algebraic inverses; this is definitional, but it is the intended reasoning semantics rather than a result derived from the evaluation. The end-to-end training does supervise the rule learner with ground-truth rule labels via Lop (Eq. 23) and Lr (Eq. 24), so the rule learner is fitted to rule annotations; however, the headline result is answer-selection accuracy on held-out RAVEN/I-RAVEN samples, which is not statistically forced by the rule-label fit because perception, rule execution, and candidate comparison must still succeed. Moreover, Section V-D reports a reasoning-only experiment (Table X) trained with the consistency loss Lrs (Eq. 27) without ground-truth rule labels, reaching 99.2% on I-RAVEN; this independently validates the rule-abduction mechanism. The Appendix A admission that the simplified logical relation function cannot express XOR and the assertion that RAVEN only uses Arithmetic+ and Arithmetic- logical rules is an empirical coverage assumption, not a circular step; it is a correctness and expressiveness risk, not a reduction of the result to its inputs. No load-bearing self-citation or imported uniqueness theorem is present: the primary cited baselines (NVSA, PrAE, ALANS, ARLC, Learn-VRF) are external, and the only overlapping-author citation ([27]) is used for a non-essential architectural observation. Therefore the paper is self-contained against external benchmarks and exhibits no significant circularity.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

Rel-SAR introduces no new physical or ontological entities. Its innovations are representational (diverse atomic vector types, SHDR, relation functions) and algorithmic (rule learners), all derived from known VSA operations. The free parameters listed are hyperparameters that affect the empirical results but are not fitted to a target quantity in the sense of a physical theory. The axioms are dominated by the dataset-specific rule vocabulary of RAVEN, which the method is explicitly designed to exploit.

free parameters (3)
  • vector dimension d = 3000
    Chosen in Section V.A for all vectors; affects capacity and pseudo-orthogonality.
  • inverse softmax temperature beta = not reported
    Introduced in Equation 15 for attention normalization; value not specified in the paper.
  • period L of circular position vectors = n^2 (4 for 2x2Grid, 9 for 3x3Grid)
    Chosen in Section IV.C.3 to match grid size; encodes cyclic shift of positions.
assumptions (5)
  • standard math FHRR/HRR binding and unbinding operations preserve structure and allow approximate retrieval
    Used throughout Section III and IV for SHDR construction and decomposition (Eq. 10-12).
  • standard math Fractional power encoding maps real addition to binding
    Equation 4 from Frady et al. [19], used for all numerical rules.
  • domain assumption RAVEN rule distribution is limited to constant, progression, arithmetic, and distribute-three over five attributes, with all numerical rules representable as additive binding powers
    Stated in Table III/IV; the design of relation functions (Eq. 6-7) and the OP/r table are built to match this exact rule set.
  • ad hoc to paper The simplified logical relation function (Eq. 7) is sufficient because RAVEN only uses Arithmetic+ and Arithmetic- for position
    Paper admits in Appendix A that the simplified function cannot express XOR; this is a dataset-specific restriction rather than a general logical reasoning capability.
  • domain assumption The perception frontend can be trained to approximate the theoretical SHDR S from images
    Training objective Eq. 13 assumes the neural encoder can minimize the distance to the hand-constructed S built from ground-truth attribute labels.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Systematic Abductive Reasoning via Diverse Relation Representations in Vector-symbolic Architecture." pith.science (2026). https://pith.science/paper/JQQADV7J

@misc{pith2026250111896,
  author       = {Pith},
  title        = {Pith review of: Systematic Abductive Reasoning via Diverse Relation Representations in Vector-symbolic Architecture},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JQQADV7J}},
  note         = {Machine review of arXiv:2501.11896}
}
read the original abstract

In abstract visual reasoning, monolithic deep learning models suffer from limited interpretability and generalization, while existing neuro-symbolic approaches fall short in capturing the diversity and systematicity of attributes and relation representations. To address these challenges, we propose a Systematic Abductive Reasoning model with diverse relation representations (Rel-SAR) in Vector-symbolic Architecture (VSA) to solve Raven's Progressive Matrices (RPM). To derive attribute representations with symbolic reasoning potential, we introduce not only various types of atomic vectors that represent numeric, periodic and logical semantics, but also the structured high-dimentional representation (SHDR) for the overall Grid component. For systematic reasoning, we propose novel numerical and logical relation functions and perform rule abduction and execution in a unified framework that integrates these relation representations. Experimental results demonstrate that Rel-SAR achieves significant improvement on RPM tasks and exhibits robust out-of-distribution generalization. Rel-SAR leverages the synergy between HD attribute representations and symbolic reasoning to achieve systematic abductive reasoning with both interpretable and computable semantics.

Figures

Figures reproduced from arXiv: 2501.11896 by the authors.

Figure 1
Figure 1. Illustrations for RAVEN dataset. (a) An example of RPM test from RAVEN [21] dataset. In an RPM test, there are 8 context panels and 8 candidate panels. Participants are required to identify the underlying rules governing various attributes within the context panels. Subsequently, participants use these rules to infer the attributes of the missing panel (represented by ”?”) and choose the most appropriate option (hig… view at source ↗
Figure 2
Figure 2. Atomic HD representations and relation functions. (a-d) The Rel￾SAR model utilizes four types of atomic HD vectors. Random Vectors (RVs), sampled independently, are used to represent distinct and unrelated symbols or concepts. Numeric Vectors (NVs) are used to represent real numbers and support VSA-based addition-type arithmetic operations. Circular Vectors (CVs) represent periodic values and enable addition-type ar… view at source ↗
Figure 3
Figure 3. Structured HD representations (SHDR) for the image panel and the nxn Grid. (a) SHDR for the image panel (Equation 10). Taking the Out-InGrid configuration as an example, an image panel contains multiple objects, each with four entity attributes: type, size, color, and existence. Through the first layer of role-filler binding, these attributes are combined to form a SHDR for each object. Additionally, at the layout l… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: The Systematic Abductive Reasoning model with diverse relation representations (Rel-SAR). (a) Overall architecture of Rel-SAR model. Our model consists of a visual perception frontend, which processes object attributes for 8 context and 8 candidate image panels in a RP…
Figure 5
Figure 5. Figure 5: Attention weights for existence on 2X2Grid. For the existence attribute, the presence weights (Wexist j (1) in Equation 15) represent the probability of an object being present at the jth position. The sum of these weights provides an approximate indication of the tota…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 33 canonical work pages

  1. [1]

    What one intelligence test measures: a theoretical account of the processing in the raven progressive matrices test

    P. A. Carpenter, M. A. Just, and P. Shell, “What one intelligence test measures: a theoretical account of the processing in the raven progressive matrices test.” Psychological review, vol. 97, no. 3, p. 404, 1990

    work page 1990

  2. [2]

    Development of abbreviated nine-item forms of the raven’s standard progressive matrices test,

    W. B. Bilker, J. A. Hansen, C. M. Brensinger, J. Richard, R. E. Gur, and R. C. Gur, “Development of abbreviated nine-item forms of the raven’s standard progressive matrices test,” Assessment, vol. 19, no. 3, pp. 354–369, 2012

    work page 2012

  3. [3]

    Abstraction and analogy-making in artificial intelligence,

    M. Mitchell, “Abstraction and analogy-making in artificial intelligence,” Annals of the New York Academy of Sciences , vol. 1505, no. 1, pp. 79–101, 2021

    work page 2021

  4. [4]

    Measuring abstract reasoning in neural networks,

    D. Barrett, F. Hill, A. Santoro, A. Morcos, and T. Lillicrap, “Measuring abstract reasoning in neural networks,” in International conference on machine learning. PMLR, 2018, pp. 511–520

    work page 2018

  5. [5]

    Learning to make analogies by contrasting abstract relational structure,

    F. Hill, A. Santoro, D. G. Barrett, A. S. Morcos, and T. Lillicrap, “Learning to make analogies by contrasting abstract relational structure,” arXiv preprint arXiv:1902.00120 , 2019

    arXiv 1902

  6. [6]

    Learning perceptual inference by contrasting,

    C. Zhang, B. Jia, F. Gao, Y . Zhu, H. Lu, and S.-C. Zhu, “Learning perceptual inference by contrasting,” Advances in neural information processing systems, vol. 32, 2019

    work page 2019

  7. [7]

    Abstract reasoning with distracting features,

    K. Zheng, Z.-J. Zha, and W. Wei, “Abstract reasoning with distracting features,” Advances in Neural Information Processing Systems , vol. 32, 2019

    work page 2019

  8. [8]

    Stratified rule-aware network for abstract visual reasoning,

    S. Hu, Y . Ma, X. Liu, Y . Wei, and S. Bai, “Stratified rule-aware network for abstract visual reasoning,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 35, no. 2, 2021, pp. 1567–1574

    work page 2021

Show all 43 references
  1. [9]

    Scale-localized abstract reasoning,

    Y . Benny, N. Pekar, and L. Wolf, “Scale-localized abstract reasoning,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 12 557–12 565

    2021

  2. [10]

    Abstract spatial-temporal reasoning via probabilistic abduction and execution,

    C. Zhang, B. Jia, S.-C. Zhu, and Y . Zhu, “Abstract spatial-temporal reasoning via probabilistic abduction and execution,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 9736–9746

    2021

  3. [11]

    G. F. Marcus, The algebraic mind: Integrating connectionism and cognitive science. MIT press, 2003

    2003

  4. [12]

    A neuro-vector-symbolic architecture for solving raven’s progressive matrices,

    M. Hersche, M. Zeqiri, L. Benini, A. Sebastian, and A. Rahimi, “A neuro-vector-symbolic architecture for solving raven’s progressive matrices,” Nature Machine Intelligence, vol. 5, no. 4, pp. 363–375, 2023

    2023

  5. [13]

    Learning algebraic representation for systematic generalization in abstract reason- ing,

    C. Zhang, S. Xie, B. Jia, Y . N. Wu, S.-C. Zhu, and Y . Zhu, “Learning algebraic representation for systematic generalization in abstract reason- ing,” in European Conference on Computer Vision. Springer, 2022, pp. 692–709

    2022

  6. [14]

    Towards learning abductive reasoning using vsa distributed representations,

    G. Camposampiero, M. Hersche, A. Terzi ´c, R. Wattenhofer, A. Se- bastian, and A. Rahimi, “Towards learning abductive reasoning using vsa distributed representations,” in International Conference on Neural- Symbolic Learning and Reasoning . Springer, 2024, pp. 370–385

    2024

  7. [15]

    Emergence of abstract rules in the primate brain,

    F. A. Mansouri, D. J. Freedman, and M. J. Buckley, “Emergence of abstract rules in the primate brain,” Nature Reviews Neuroscience , vol. 21, no. 11, pp. 595–610, 2020

    2020

  8. [16]

    Insights for ai from the human mind,

    G. Marcus and E. Davis, “Insights for ai from the human mind,” Communications of the ACM , vol. 64, no. 1, pp. 38–41, 2020

    2020

  9. [17]

    The empirical case for two systems of reasoning

    S. A. Sloman, “The empirical case for two systems of reasoning.” Psychological bulletin, vol. 119, no. 1, p. 3, 1996

    1996

  10. [18]

    T. A. Plate, Distributed representations and nested compositional struc- ture. Citeseer, 1994

    1994

  11. [19]

    Com- puting on functions using randomized vector representations

    E. Frady, D. Kleyko, C. Kymn, B. Olshausen, and F. Sommer, “Com- puting on functions using randomized vector representations.” arXiv: Learning,arXiv: Learning, Sep 2021

    2021

  12. [20]

    A survey on hyperdimensional computing aka vector symbolic architectures, part i: Models and data transformations,

    D. Kleyko, D. A. Rachkovskij, E. Osipov, and A. Rahimi, “A survey on hyperdimensional computing aka vector symbolic architectures, part i: Models and data transformations,” ACM Computing Surveys, vol. 55, no. 6, pp. 1–40, 2022

    2022

  13. [21]

    Raven: A dataset for relational and analogical visual reasoning,

    C. Zhang, F. Gao, B. Jia, Y . Zhu, and S.-C. Zhu, “Raven: A dataset for relational and analogical visual reasoning,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 5317–5327

    2019

  14. [22]

    A simple neural network module for re- lational reasoning,

    A. Santoro, D. Raposo, D. G. Barrett, M. Malinowski, R. Pascanu, P. Battaglia, and T. Lillicrap, “A simple neural network module for re- lational reasoning,” Advances in neural information processing systems , vol. 30, 2017

    2017

  15. [23]

    Solving raven’s progressive matrices with multi-layer relation networks,

    M. Jahrens and T. Martinetz, “Solving raven’s progressive matrices with multi-layer relation networks,” in 2020 International Joint Conference on Neural Networks (IJCNN) . IEEE, 2020, pp. 1–6

    2020

  16. [24]

    Multi-label contrastive learning for abstract visual reasoning,

    M. Małki ´nski and J. Ma ´ndziuk, “Multi-label contrastive learning for abstract visual reasoning,” IEEE Transactions on Neural Networks and Learning Systems, vol. 35, no. 2, pp. 1941–1953, 2022

    1941

  17. [25]

    Effective abstract reasoning with dual- contrast network,

    T. Zhuo and M. Kankanhalli, “Effective abstract reasoning with dual- contrast network,” arXiv preprint arXiv:2205.13720 , 2022

    2022 arXiv

  18. [26]

    Learning visual abstract reasoning through dual-stream networks,

    K. Zhao, C. Xu, and B. Si, “Learning visual abstract reasoning through dual-stream networks,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 15, 2024, pp. 16 979–16 988

    2024

  19. [27]

    Neural prediction errors enable analogical visual reasoning in human standard intelligence tests,

    L. Yang, H. You, Z. Zhen, D. Wang, X. Wan, X. Xie, and R.-Y . Zhang, “Neural prediction errors enable analogical visual reasoning in human standard intelligence tests,” in International Conference on Machine Learning. PMLR, 2023, pp. 39 572–39 583

    2023

  20. [28]

    The scattering compositional learner: Discovering objects, attributes, relationships in analogical rea- soning,

    Y . Wu, H. Dong, R. Grosse, and J. Ba, “The scattering compositional learner: Discovering objects, attributes, relationships in analogical rea- soning,” arXiv preprint arXiv:2007.04212 , 2020

    2007 arXiv

  21. [29]

    Learning to reason over visual objects,

    S. S. Mondal, T. Webb, and J. D. Cohen, “Learning to reason over visual objects,” arXiv preprint arXiv:2303.02260 , 2023

    2023 arXiv

  22. [30]

    A closer look at generalisation in raven,

    S. Spratley, K. Ehinger, and T. Miller, “A closer look at generalisation in raven,” in Computer Vision–ECCV 2020: 16th European Confer- ence, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXVII 16 . Springer, 2020, pp. 601–616

    2020

  23. [31]

    Probabilistic abduction for visual abstract reasoning via learning rules in vector-symbolic architectures,

    M. Hersche, F. Di Stefano, T. Hofmann, A. Sebastian, and A. Rahimi, “Probabilistic abduction for visual abstract reasoning via learning rules in vector-symbolic architectures,” arXiv preprint arXiv:2401.16024 , 2024

    2024 arXiv

  24. [32]

    Emergent symbols through binding in external memory,

    T. W. Webb, I. Sinha, and J. D. Cohen, “Emergent symbols through binding in external memory,” arXiv preprint arXiv:2012.14601 , 2020

    2012 arXiv

  25. [33]

    Abstractors and rela- tional cross-attention: An inductive bias for explicit relational reasoning in transformers,

    A. Altabaa, T. Webb, J. Cohen, and J. Lafferty, “Abstractors and rela- tional cross-attention: An inductive bias for explicit relational reasoning in transformers,” arXiv preprint arXiv:2304.00195 , 2023. 15

    2023 arXiv

  26. [34]

    On neural architecture inductive biases for relational tasks. arxiv,

    G. Kerg, S. Mittal, D. Rolnick, Y . Bengio, B. Richards, and G. Lajoie, “On neural architecture inductive biases for relational tasks. arxiv,”arXiv preprint arXiv:2206.05056, 2022

    2022 arXiv

  27. [35]

    The relational bottleneck as an inductive bias for efficient abstraction,

    T. W. Webb, S. M. Frankland, A. Altabaa, S. Segert, K. Krishnamurthy, D. Campbell, J. Russin, T. Giallanza, R. O’Reilly, J. Lafferty et al. , “The relational bottleneck as an inductive bias for efficient abstraction,” Trends in Cognitive Sciences , 2024

    2024

  28. [36]

    Holographic reduced representations,

    T. A. Plate, “Holographic reduced representations,” IEEE Transactions on Neural networks , vol. 6, no. 3, pp. 623–641, 1995

    1995

  29. [37]

    Hyperdimensional computing: An introduction to comput- ing in distributed representation with high-dimensional random vectors,

    P. Kanerva, “Hyperdimensional computing: An introduction to comput- ing in distributed representation with high-dimensional random vectors,” Cognitive computation, vol. 1, pp. 139–159, 2009

    2009

  30. [38]

    Attention is all you need,

    A. Vaswani, “Attention is all you need,” Advances in Neural Information Processing Systems, 2017

    2017

  31. [39]

    Abstract diagrammatic reasoning with multiplex graph networks,

    D. Wang, M. Jamnik, and P. Lio, “Abstract diagrammatic reasoning with multiplex graph networks,” arXiv preprint arXiv:2006.11197 , 2020

    2006 arXiv

  32. [40]

    Abstract visual reasoning: An algebraic approach for solving raven’s progressive matrices,

    J. Xu, T. Vaidya, Y . Wu, S. Chandra, Z. Lai, and K. F. E. Chong, “Abstract visual reasoning: An algebraic approach for solving raven’s progressive matrices,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023, pp. 6715–6724

    2023

  33. [41]

    Language models are few-shot learners,

    T. B. Brown, “Language models are few-shot learners,” arXiv preprint arXiv:2005.14165, 2020

    2005 arXiv

  34. [42]

    Object-centric learning with slot attention,

    F. Locatello, D. Weissenborn, T. Unterthiner, A. Mahendran, G. Heigold, J. Uszkoreit, A. Dosovitskiy, and T. Kipf, “Object-centric learning with slot attention,” Advances in neural information processing systems , vol. 33, pp. 11 525–11 538, 2020

    2020

  35. [43]

    Neural discrete representation learning,

    A. Van Den Oord, O. Vinyals et al. , “Neural discrete representation learning,” Advances in neural information processing systems , vol. 30, 2017. APPENDIX A THE FULL VERSION OF LOGICAL RELATION FUNCTION The RPM-style logic rules are expressed as that the set of attribute valu...

    2017

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.