Pith. sign in

REVIEW 4 major objections 4 minor 42 references

sLTN: Structural Logic Tensor Networks

T0 review · 4 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read sLTN extends Logic Tensor Networks with structural dimensions, named tensor axes for time, sequence position, or graph nodes, so logical formulas can quantify over structure directly and train with differentiable fuzzy semantics; without…

desk verdict A genuine formal extension of LTN for structured data with a working library; the main soft spots are the unproved recovery claim, a typo in one guarded-quantification rule, and the fixed-finite-extent assumption that limits the expressiveness claim. read the letter →

arxiv 2608.11136 v1 pith:FOFBIOMZ submitted 2026-08-11 cs.AI

Davide Rinaldi , Luciano Serafini This is my paper · ORCID
classification cs.AI
keywords logictensornetworksneurosymbolicreasoningstructuraldimensionsfuzzysemanticstemporalgraphrelationsdifferentiablelearningtypedlogicalsignatures
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

Logic Tensor Networks treat data as flat collections of individuals, which makes temporal order, sequence position, and graph adjacency hard to state as logical conditions. sLTN makes structural organization part of the language itself: structural dimensions are named tensor axes for time steps, sequence positions, or graph nodes, and structural relations such as next(t,t1) are grounded as fuzzy masks. Formulas can quantify over these axes, guard universal statements with adjacency masks, select slices, and align expressions by axis name, while retaining the differentiable fuzzy semantics of LTN. If correct, this gives a single framework where learning and temporal or relational logical reasoning happen in the same gradient-based objective, and standard LTN becomes the special case with no structural dimensions.

What carries the argument

The load-bearing mechanism is the annotated tensor with three kinds of axes, namely variable axes for individuals, structural axes for positions such as time steps or graph nodes, and domain axes for feature dimensions, together with structural relations, which are Boolean or fuzzy masks over structural axes. Structural variables range over the indices of a declared dimension, guarded quantification uses a relation's mask to restrict or weight an aggregation, and axis selection fixes a position such as the first or last time step. The structural profile of each function and predicate declares which axes it consumes and produces, so a frame-level classifier can be applied pointwise along a temporal axis while a sequence-level predicate consumes that axis. This axis machinery carries the entire claim because it is what makes temporal persistence, adjacency, and slice conditions expressible as differentiable logical clauses.

What would settle it

Take a minibatch of videos with different lengths, pad them to a common temporal extent, and evaluate the persistence clause $\psi$ from the paper, which states that completeness at time $t$ implies completeness at the next time. Because the next mask marks the boundary between the last real frame and a padded frame as an adjacency, the formula can fail on padding artifacts even though the underlying sequence satisfies persistence; this would demonstrate that the semantics depends on the fixed-extent encoding as much as on the data's actual order.

Watch

Extended reading notes

Core claim

The central claim is that positional and relational organization can be moved from data preprocessing into the logical language. sLTN declares structural dimensions in the signature, lets variables carry them, consumes them through predicates such as a whole-sequence classifier, or propagates them pointwise through a frame-level classifier, and expresses relations such as 'next' as masks over axes. Every term and formula is denoted by an annotated tensor whose axes are explicitly labeled as variable, structural, or domain, and all composition is defined by structural recursion with named-axis alignment and broadcasting. Learning is formulated as maximizing satisfiability of a knowledge base, either by aggregating clause truth values or by combining per-clause gradients. In the absence of declared structural dimensions, the denotations described in the paper reduce exactly to standard LTN semantics.

Load-bearing premise

The whole construction assumes that every structural dimension has a fixed finite extent, a preset number of time steps, positions, or nodes, in each evaluation context; variable-length or irregularly connected data must be padded or encoded into that fixed grid, and structural relations must be supplied as masks.

Editorial extensions

If this is right

  • A temporal persistence law such as 'if the digit is complete at time $t$, it is complete at the next time' becomes a guarded structural formula that can be trained by backpropagation.
  • Graph or sequence adjacency can be imposed as a fuzzy mask without materializing positions as first-order individuals, so large structural axes do not need their own feature vectors.
  • The framework is conservative: a theory with no structural dimensions has exactly the standard LTN semantics, so existing flat LTN formulations fit unchanged.
  • Knowledge-base clauses may be kept as separate objectives and combined by Jacobian descent, allowing conflicting clauses to be handled beyond plain scalarization.
  • Signature declarations and parsing make arity, sort, and dimension errors detectable before grounding, instead of only when tensor shapes fail to match.

Reading between the lines

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

  • The paper only illustrates temporal and sequential examples, but the same mechanism directly suggests graph reasoning: declare nodes as a structural dimension, ground adjacency as a mask, and write message-passing-style constraints as guarded quantifiers; testing this on node classification would extend the claim beyond the running example.
  • Because structural relations are just masks, learned masks would let the framework discover adjacency or transition structure from data while still preserving logical interpretation; the paper does not explore this, but it follows from the semantics.
  • The reduction to plain LTN implies that existing LTN applications should run without modification under sLTN when no structural dimensions are declared; a benchmark regression across prior LTN tasks would verify this compatibility claim.
  • The distinction between consuming and propagating structural axes is what lets frame-level and sequence-level predicates coexist; one could test its limits by mixing several structural dimensions, such as time and space, in one formula, which the current examples do not do.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The paper proposes sLTN, an extension of Logic Tensor Networks in which structural dimensions—temporal, sequential, positional, or graph-like axes—are made first-class syntactic entities. It defines a many-sorted signature with structural dimension profiles, a term/formula grammar with structural variables, structural relations, axis renaming, axis slicing, and guarded structural quantification, and a fuzzy tensor semantics in which expressions denote annotated tensors with named variable, structural, and domain axes. The paper states that sLTN reduces to standard LTN when no structural dimensions are declared, describes a PyTorch implementation with a parser/signature separation and both scalarized and multi-objective training, and illustrates the framework on a synthetic video-classification example.

Significance. If its claims are secured, the paper makes a useful conceptual and engineering contribution: it provides a typed, declarative syntax and a compositional tensor semantics for making positional and relational organization explicit in differentiable neuro-symbolic reasoning, together with a companion library. The formal notation is detailed, the running example is internally consistent, and the implementation section gives concrete grounding for the definitions. The paper is also appropriately modest about empirical validation, explicitly stating in Section 6 that systematic experiments are future work. The main reservations concern the scope of the expressiveness claim under fixed finite extents, the absence of a formal proof of the claimed reduction to LTN, and a typo in the guarded structural quantification rule; these are load-bearing for the paper's central contribution and need to be addressed.

major comments (4)
  1. [Section 2.3.2(v)] The formal rule for guarded structural quantification contains an undefined symbol and a misaligned operand. The text defines align(Ipϕq,Ipρq) as p zIpρq, zIpσqq, but σ has not been introduced; it should be align(Ipϕq,Ipρq) = (zIpϕq, zIpρq). The subsequent aggregation uses vppρq, which should be vpzIpρq after alignment. Since guarded structural quantification is one of the key new constructs, this rule must be corrected.
  2. [Section 2.3.1] The semantics assumes every structural dimension has a fixed finite extent in the 'current evaluation context', but the evaluation context itself is never defined, nor is there a consistency condition for occurrences of the same dimension within one formula. For variable-length sequences or graphs whose node sets change, the user must pad or clip to a fixed extent and supply structural relations as masks, but the paper does not analyze how padding interacts with guarded aggregation, axis slicing, or the requirement that occurrences of the same dimension have compatible extents. The central claim that temporal, sequential, and graph structure is made first-class is therefore only supported for structures of a pre-chosen cardinality; the paper should either state this limitation explicitly and discuss its consequences, or extend the semantics to handle variable extents.
  3. [Abstract and Section 6] The paper claims that sLTN 'recovers the original LTN semantics as a special case' when no structural dimensions are declared, but no theorem or proof of this reduction is provided. The claim appears plausible by inspection, but it is a compatibility guarantee that is load-bearing for the framing of the contribution. Add a formal proposition and a proof sketch showing that, when all dimensional profiles are empty and no structural quantifiers or structural relations occur, the denotation rules of Section 2.3 coincide with standard LTN semantics.
  4. [Sections 2.1 and 2.3.2(iii)] The rule for consuming structural axes in function and predicate application is underspecified when several arguments carry axes of the same declared dimension. Section 2.1 says that matching is performed 'position-wise according to the ordered profile' dims_in(g), and Section 2.3.2(iii) says that all aligned structural axes whose dimensions occur in the input profile are consumed, but these two statements are not reconciled and no formal definition determines which argument axes are consumed when, for example, two arguments both carry a temporal axis. Provide an explicit technical condition for the consumption relation.
minor comments (4)
  1. [Section 2.3.2(vi)] In the diagonal quantification rule, the text refers to diag_{x1,...,xh}(vpθq) and diag_{x1,...,xh}(vpϕq) in the same paragraph; the first occurrence should be vpϕq.
  2. [Example 2] Formula β contains the token l_T in the concrete string 'Complete(x[t][t=l_T-1])', but l_T is not declared in the signature and no substitution mechanism for such a placeholder is described. Clarify whether l_T is a Python variable inserted into the string or a signature constant.
  3. [Section 4] The sentence 'is available at available at https://...' contains a duplicated phrase; remove the repetition.
  4. [Section 2.2] The paper states that expressions must satisfy typing constraints but does not define a formal well-formedness relation; the phrase 'well formed' is used throughout the semantics without a precise inductive definition. A short formal definition would make the dependency of the semantics on syntactic validity explicit.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: sLTN's structural semantics are independently defined; the LTN-recovery claim is a definitional special case, not a fitted prediction or a self-citation-forced conclusion.

full rationale

The paper's central contribution is a formal language extension, not an empirical prediction: it defines structural dimensions, structural variables, and structural relations, and gives a compositional tensor semantics. Nothing in the derivation is fitted to data and then renamed as a prediction; the only "recovery" claim, that sLTN reduces to LTN when no structural dimensions are declared, follows directly from the definitions (empty dimensional profiles leave the groundings and denotation rules in their standard LTN form), and the paper presents it as a compatibility statement rather than as an independently discovered result. This is a definitional special case, not a circular derivation. The self-citations to the authors' prior LTN papers ([1], [34], [9]) supply background and operator conventions, but no load-bearing uniqueness theorem or ansatz is imported from them to force the current framework's choices; the new structural semantics is specified within this paper with explicit grammar, typing, and denotation rules. The fixed finite extent assumption in Section 2.3.1 is a stated modeling limitation for variable-length or irregular structures, and the paper itself acknowledges in Section 6 that systematic empirical validation is still needed; these are scope and correctness risks, not circularity. Overall, the derivation chain is self-contained and no circular step is present.

Assumptions & free parameters 4 free parameters · 5 assumptions · 2 invented entities

The ledger is modest for a framework paper: the main free choices are fuzzy aggregation exponents and a stabilization offset, and the load-bearing domain assumptions are finite extents and user-supplied structural masks. The invented entities are representational constructs inside the library rather than empirical postulates, so they carry no independent falsifiable evidence.

free parameters (4)
  • universal aggregation exponent p (global) = p=1/5 in running example
    Set by hand in Section 2.3.1 for interp.logic['forall']; controls how strictly universal truth is aggregated and changes gradient flow.
  • structural universal aggregation exponent p (temporal axis) = p=10 in running example
    Set by hand in Section 2.3.1 for interp.logic['forall,T']; chosen without a stated selection criterion and affects the temporal quantifier semantics.
  • default power-mean and SatAgg exponents = p=2 (default)
    Appendix A defines AggregPMean and AggregPMeanError with default p=2, including satisfaction aggregation, so the exact learning objective depends on these hand-chosen defaults.
  • stabilization offset epsilon = not specified
    Appendix A.3 defines the pi0 and pi1 projections using an epsilon offset but never gives a numerical value, leaving numerical results not fully pinned down.
assumptions (5)
  • standard math Many-sorted first-order logic with guarded and diagonal quantification is taken as the base language.
    Section 2.2 defines sLTN formulas on top of this base; the recovery claim relies on the LTN semantics of [1] for the base case.
  • standard math Logical connectives and aggregators are fuzzy operators satisfying Boolean boundary conditions and mapping [0,1] inputs to [0,1] outputs.
    Appendix A formalizes negation, t-norms, t-conorms, implications, and power-mean aggregators used in Section 2.3. No proof is given that every admissible configuration preserves the required properties.
  • domain assumption Every structural dimension has a finite extent and is indexed by natural numbers in the current evaluation context.
    Section 2.3.1 grounds constants and variables over extents l_d in N; this excludes unbounded or ragged structures without padding and is the load-bearing modeling assumption.
  • domain assumption Structural relations are represented by fixed Boolean or fuzzy masks over tuples of structural indices, supplied by the user.
    Section 2.3.1 grounds R as a tensor in [0,1]^{l_d1 x ... x l_dk}; the framework does not learn these masks or derive them from data.
  • domain assumption The groundings of functions, predicates, and fuzzy operators are differentiable where gradient learning is used.
    Section 3 optimizes satisfaction via gradient descent, so the chosen neural groundings and operator stabilizations must be differentiable.
invented entities (2)
  • structural dimension
    purpose: A named tensor axis used as a domain of quantification over positions such as time, sequence, or graph nodes.
    The paper adds structural dimensions as first-class logical objects; they are implementable in the library, but no externally falsifiable empirical prediction is attached to them.
  • structural relation mask
    purpose: A Boolean or fuzzy mask over tuples of structural indices, used to express relations such as next(t,t1).
    This is a new semantic category whose meaning is fully supplied by user tensors, not by an independent data source or a learned prediction with testable consequences.

how reviews work

0 comments
Cite this review

Pith. "Pith review of sLTN: Structural Logic Tensor Networks." pith.science (2026). https://pith.science/paper/FOFBIOMZ

@misc{pith2026260811136,
  author       = {Pith},
  title        = {Pith review of: sLTN: Structural Logic Tensor Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FOFBIOMZ}},
  note         = {Machine review of arXiv:2608.11136}
}
read the original abstract

Logic Tensor Networks (LTN) provide a neurosymbolic framework in which first-order logic is interpreted through tensor operations, enabling logical constraints to be integrated with differentiable learning. However, the original formulation of LTN is primarily suited to data represented as flat collections of individuals, and does not explicitly capture structural organization such as temporal order, sequential position, or graph connectivity. We introduce sLTN, an extension of LTN that makes structural dimensions first-class elements of the language. Structural dimensions represent named tensor axes associated with domain-specific organization, such as time steps, sequence positions, or graph nodes. They can be quantified explicitly, related through structural relations, and used to express temporal, sequential, and relational constraints directly at the logical level. We formalize the syntax and fuzzy tensor semantics of sLTN and show that, in the absence of structural dimensions, the framework recovers the original LTN semantics as a special case. We further describe a PyTorch implementation based on a declarative signature, formula parsing, and tensorial interpretation. The framework is illustrated on representative temporal and sequential reasoning examples. This paper serves as a companion to the sltn library, available at https://github.com/logictensornetworks/sltn.

Figures

Figures reproduced from arXiv: 2608.11136 by the authors.

Figure 1
Figure 1. Running video-classification example: a sequence of frames shows a handwritten digit [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Crisp, left, and soft, right, interpretations of the [PITH_FULL_IMAGE:figures/full_fig_p015_2.png] view at source ↗
Figure 3
Figure 3. Interpretation of the atomic formula ppx, fpx, yqq, adapted from [1]. 19 [PITH_FULL_IMAGE:figures/full_fig_p019_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Spectrum of the power-mean aggregators across exponents [PITH_FULL_IMAGE:figures/full_fig_p036_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 36 canonical work pages

  1. [1]

    Logic tensor networks.Artificial Intelligence, 303:103649, 2022

    Samy Badreddine, Artur d’Avila Garcez, Luciano Serafini, and Michael Spranger. Logic tensor networks.Artificial Intelligence, 303:103649, 2022

    work page 2022

  2. [2]

    Jonathan T. Barron. Squareplus: A softplus-like algebraic rectifier.CoRR, abs/2112.11687, 2021

    arXiv 2021

  3. [3]

    Integrating background knowledge in medical semantic segmentation with logic tensor networks

    Luca Bergamin, Giovanna Maria Dimitri, and Fabio Aiolli. Integrating background knowledge in medical semantic segmentation with logic tensor networks. In2025 International Joint Conference on Neural Networks (IJCNN), pages 1–7. IEEE, 2025

    work page 2025

  4. [4]

    First-Order Temporal Logic Tensor Networks

    Luca Boscarato, Ivan Donadello, Alessandro Artale, Marco Montali, and Fabrizio Maria Maggi. First-order temporal logic tensor networks.arXiv preprint arXiv:2606.29972, 2026

    work page Pith review arXiv 2026

  5. [5]

    Enhancing multi-label learning in visual scenes with logic tensor networks and positive-unlabeled learning.SSRN Electronic Journal, 2025

    Youssef Bouaziz and Vincent Barra. Enhancing multi-label learning in visual scenes with logic tensor networks and positive-unlabeled learning.SSRN Electronic Journal, 2025

    work page 2025

  6. [6]

    Springer Science & Business Media, 2013

    Peter S Bullen.Handbook of means and their inequalities, volume 560. Springer Science & Business Media, 2013

    work page 2013

  7. [7]

    Overcoming recommendation limitations with neuro-symbolic integration

    Tommaso Carraro. Overcoming recommendation limitations with neuro-symbolic integration. InProceedings of the 17th ACM Conference on Recommender Systems, pages 1325–1331, 2023. 28

    work page 2023

  8. [8]

    Mitigating data sparsity via neuro-symbolic knowledge transfer

    Tommaso Carraro, Alessandro Daniele, Fabio Aiolli, and Luciano Serafini. Mitigating data sparsity via neuro-symbolic knowledge transfer. InEuropean Conference on Information Re- trieval, pages 226–242. Springer, 2024

    work page 2024

Show all 42 references
  1. [9]

    Ltntorch: Pytorch implementation of logic tensor networks.arXiv preprint arXiv:2409.16045, 2024

    Tommaso Carraro, Luciano Serafini, and Fabio Aiolli. Ltntorch: Pytorch implementation of logic tensor networks.arXiv preprint arXiv:2409.16045, 2024

    2024 arXiv

  2. [10]

    Weakly supervised segmenta- tion as semantic-based regularization.arXiv preprint arXiv:2605.13674, 2026

    Stefano Colamonaco, Andrei-Bogdan Florea, and Jaron Maene. Weakly supervised segmenta- tion as semantic-based regularization.arXiv preprint arXiv:2605.13674, 2026

    2026 arXiv

  3. [11]

    Large language model enhanced logic tensor network for stance detection.Neural Networks, 183:106956, 2025

    Genan Dai, Jiayu Liao, Sicheng Zhao, Xianghua Fu, Xiaojiang Peng, Hu Huang, and Bowen Zhang. Large language model enhanced logic tensor network for stance detection.Neural Networks, 183:106956, 2025

    2025

  4. [12]

    Knowledge enhanced neural networks

    Alessandro Daniele and Luciano Serafini. Knowledge enhanced neural networks. InPacific Rim International Conference on Artificial Intelligence, pages 542–554. Springer, 2019

    2019

  5. [13]

    Enhancing logical tensor networks: Integrating uninorm-based fuzzy operators for complex reasoning

    Paulo Vitor de Campos Souza, Gianluca Apriceno, and Mauro Dragoni. Enhancing logical tensor networks: Integrating uninorm-based fuzzy operators for complex reasoning. InInter- national Conference on Neural-Symbolic Learning and Reasoning, pages 68–79. Springer, 2024

    2024

  6. [14]

    Neuro-symbolic learning for pre- dictiveprocessmonitoringviatwo-stagelogictensornetworkswithrulepruning

    Fabrizio De Santis, Gyunam Park, and Francesco Zanichelli. Neuro-symbolic learning for pre- dictiveprocessmonitoringviatwo-stagelogictensornetworkswithrulepruning. InPacific-Asia Conference on Knowledge Discovery and Data Mining, pages 104–118. Springer, 2026

    2026

  7. [15]

    Neurosymbolic visual transform based on logic tensor network for defect detection

    Youcef Djenouri, Ahmed Nabil Belbachir, Asma Belhadi, and Tomasz Michalak. Neurosymbolic visual transform based on logic tensor network for defect detection. InEuropean Conference on Computer Vision, pages 18–34. Springer, 2024

    2024

  8. [16]

    Compensating supervision incompleteness with prior knowledge in semantic image interpretation

    Ivan Donadello and Luciano Serafini. Compensating supervision incompleteness with prior knowledge in semantic image interpretation. In2019 International Joint Conference on Neural Networks (IJCNN), pages 1–8. IEEE, 2019

    2019

  9. [17]

    Logic tensor networks for semantic image interpretation

    Ivan Donadello, Luciano Serafini, and Artur d’Avila Garcez. Logic tensor networks for semantic image interpretation. InProceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, IJCAI-17, pages 1596–1602, 2017

    2017

  10. [18]

    Extending logic tensor networks to im- plicit feedback for representation-aware music recommendation

    Hannah Eckert, Oleg Lesota, and Markus Schedl. Extending logic tensor networks to im- plicit feedback for representation-aware music recommendation. InEuropean Conference on Information Retrieval, pages 428–441. Springer, 2026

    2026

  11. [19]

    Neuro-symbolic process anomaly detection

    Devashish Gaikwad1, Wil MP van der Aalst1, and Gyunam Park. Neuro-symbolic process anomaly detection. InAdvanced Information Systems Engineering: 38th International Confer- ence, CAiSE 2026, Verona, Italy, June 8–12, 2026, Proceedings, Part II, page 240. Springer Nature, 2026

    2026

  12. [20]

    Enhancing transcription factor prediction via domainknowledgeintegrationwithlogictensornetworks.IEEE Transactions on Computational Biology and Bioinformatics, 2025

    Liyuan Gao, Linpeng Sun, and Victor S Sheng. Enhancing transcription factor prediction via domainknowledgeintegrationwithlogictensornetworks.IEEE Transactions on Computational Biology and Bioinformatics, 2025

    2025

  13. [21]

    From large language model predicates to logic tensor networks: Neurosymbolic offer validation in regulated procurement

    Cedric S Haufe and Frieder Stolzenburg. From large language model predicates to logic tensor networks: Neurosymbolic offer validation in regulated procurement. InGerman Conference on Artificial Intelligence (Künstliche Intelligenz), pages 236–243. Springer, 2026. 29

    2026

  14. [22]

    Harnessing deep neural networks with logic rules

    Zhiting Hu, Xuezhe Ma, Zhengzhong Liu, Eduard Hovy, and Eric Xing. Harnessing deep neural networks with logic rules. InProceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2410–2420, Berlin, Germany, August 2016...

    2016

  15. [23]

    Deepproblog: Neural probabilistic logic programming

    Robin Manhaeve, Sebastijan Dumancic, Angelika Kimmig, Thomas Demeester, and Luc De Raedt. Deepproblog: Neural probabilistic logic programming. InProceedings of the 32nd International Conference on Neural Information Processing Systems, NeurIPS’18, pages 3753– 3763, USA, 2018. ...

    2018

  16. [24]

    Boosting zero-shot learning through neuro-symbolic integration.Pattern Recognition, 170:111869, 2026

    Francesco Manigrasso, Fabrizio Lamberti, and Lia Morra. Boosting zero-shot learning through neuro-symbolic integration.Pattern Recognition, 170:111869, 2026

    2026

  17. [25]

    Probing llms for logical reasoning

    Francesco Manigrasso, Stefan Schouten, Lia Morra, and Peter Bloem. Probing llms for logical reasoning. InInternational conference on neural-symbolic learning and reasoning, pages 257–

  18. [26]

    Integrating learning and reasoning with deep logic models

    Giuseppe Marra, Francesco Giannini, Michelangelo Diligenti, and Marco Gori. Integrating learning and reasoning with deep logic models. InMachine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2019, Würzburg, Germany, September 16-20, 2019, Proce...

    2019

  19. [27]

    Lyrics: A gen- eral interface layer to integrate logic inference and deep learning

    Giuseppe Marra, Francesco Giannini, Michelangelo Diligenti, and Marco Gori. Lyrics: A gen- eral interface layer to integrate logic inference and deep learning. InJoint European Conference on Machine Learning and Knowledge Discovery in Databases, pages 283–298. Springer, 2019

    2019

  20. [28]

    A logic tensor network-based neurosymbolic framework for explainable diabetes prediction.Applied Sciences, 15(21):11806, 2025

    Semanto Mondal, Antonino Ferraro, Fabiano Pecorelli, and Giuseppe De Pietro. A logic tensor network-based neurosymbolic framework for explainable diabetes prediction.Applied Sciences, 15(21):11806, 2025

    2025

  21. [29]

    On the t-transitivity of kernels.Fuzzy Sets and Systems, 157(13):1787–1796, 2006

    Bernhard Moser. On the t-transitivity of kernels.Fuzzy Sets and Systems, 157(13):1787–1796, 2006

    2006

  22. [30]

    Pytorch: An imperative style, high-performance deep learning library, 2019

    Adam Paszke et al. Pytorch: An imperative style, high-performance deep learning library, 2019

    2019

  23. [31]

    Jacobian descent for multi-objective optimization.arXiv preprint arXiv:2406.16232, 2024

    Pierre Quinton and Valérian Rey. Jacobian descent for multi-objective optimization.arXiv preprint arXiv:2406.16232, 2024

    2024 arXiv

  24. [32]

    Markov logic networks.Mach

    Matthew Richardson and Pedro Domingos. Markov logic networks.Mach. Learn., 62(1-2):107– 136, February 2006

    2006

  25. [33]

    Logical Neural Networks.arXiv:2006.13155 [cs], June 2020

    Ryan Riegel, Alexander Gray, Francois Luus, Naweed Khan, Ndivhuwo Makondo, Ismail Yunus Akhalwaya, Haifeng Qian, Ronald Fagin, Francisco Barahona, Udit Sharma, Shajith Ikbal, Hima Karanam, Sumit Neelam, Ankita Likhyani, and Santosh Srivastava. Logical Neural Networks.arXiv:200...

    2006 arXiv

  26. [34]

    Learning and reasoning with logic tensor networks

    Luciano Serafini and Artur d’Avila Garcez. Learning and reasoning with logic tensor networks. InConference of the Italian Association for Artificial Intelligence, pages 334–348. Springer, 2016. 30

    2016

  27. [35]

    Lifted relational neural networks: Efficient learning of latent relational structures.Journal of Artificial Intelligence Research, 62:69–100, 2018

    Gustav Sourek, Vojtech Aschenbrenner, Filip Zelezny, Steven Schockaert, and Ondrej Kuzelka. Lifted relational neural networks: Efficient learning of latent relational structures.Journal of Artificial Intelligence Research, 62:69–100, 2018

    2018

  28. [36]

    Ontomedrec: Logically-pretrained model-agnostic ontology encoders for medication recommen- dation.World Wide Web, 27(3):28, 2024

    Weicong Tan, Weiqing Wang, Xin Zhou, Wray Buntine, Gordon Bingham, and Hongzhi Yin. Ontomedrec: Logically-pretrained model-agnostic ontology encoders for medication recommen- dation.World Wide Web, 27(3):28, 2024

    2024

  29. [37]

    Grounding ltlf specifications in image sequences

    Elena Umili, Roberto Capobianco, and Giuseppe De Giacomo. Grounding ltlf specifications in image sequences. InProceedings of the International Conference on Principles of Knowledge Representation and Reasoning, pages 668–678, 2023

    2023

  30. [38]

    Logictensornetwork-enhancedgenerativeadversarialnetwork

    NijeshUpretiandVaishakBelle. Logictensornetwork-enhancedgenerativeadversarialnetwork. arXiv preprint arXiv:2601.03839, 2026

    2026

  31. [39]

    Analyzing Differentiable Fuzzy Logic Operators.arXiv:2002.06100 [cs], February 2020

    Emile van Krieken, Erman Acar, and Frank van Harmelen. Analyzing Differentiable Fuzzy Logic Operators.arXiv:2002.06100 [cs], February 2020. arXiv: 2002.06100

    2002 arXiv

  32. [40]

    A Semantic Loss Function forDeepLearningwithSymbolicKnowledge

    Jingyi Xu, Zilu Zhang, Tal Friedman, Yitao Liang, and Guy Broeck. A Semantic Loss Function forDeepLearningwithSymbolicKnowledge. InInternational Conference on Machine Learning, pages 5502–5511. PMLR, July 2018. ISSN: 2640-3498

    2018

  33. [41]

    Fedltn-cubesat: Neuro-symbolic federated learning for intrusion detection in leo cubesat constellations.Mathematics, 14(6), 2026

    Gang Yang, Lin Ni, Junfeng Geng, and Xiang Peng. Fedltn-cubesat: Neuro-symbolic federated learning for intrusion detection in leo cubesat constellations.Mathematics, 14(6), 2026

    2026

  34. [42]

    Neuro-symbolic artificial intelligence: Towards improving the reasoning abilities of large language models.arXiv preprint arXiv:2508.13678, 2025

    Xiao-Wen Yang, Jie-Jing Shao, Lan-Zhe Guo, Bo-Wen Zhang, Zhi Zhou, Lin-Han Jia, Wang- Zhou Dai, and Yu-Feng Li. Neuro-symbolic artificial intelligence: Towards improving the reasoning abilities of large language models.arXiv preprint arXiv:2508.13678, 2025. A Fuzzy Operators a...

    2025 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.