Answer Set Programming Energised! End-to-End Neurosymbolic Reasoning and Learning with ASP and Energy Based Models

Jakob Suchan; Julius Monsen; Mehul Bhatt; Salim Baloch

REVIEW 3 major objections 4 minor 31 references

ASP stable models become preferred energy minima so non-monotonic logic can train neural perception end-to-end.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-10 12:30 UTC pith:5SWTBTXU

load-bearing objection Clean ASP-EBM integration that works as a platform, but the full end-to-end free-energy claim is only shown on the toy case. the 3 major comments →

arxiv 2607.08136 v1 pith:5SWTBTXU submitted 2026-07-09 cs.AI

Answer Set Programming Energised! End-to-End Neurosymbolic Reasoning and Learning with ASP and Energy Based Models

Jakob Suchan , Julius Monsen , Salim Baloch , Mehul Bhatt This is my paper

classification cs.AI

keywords answer set programmingenergy-based modelsneurosymbolic AIstable modelsend-to-end learningvisual question answeringmulti-object trackingnon-monotonic reasoning

verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper presents ASPEn, a method that fuses answer set programming with energy-based models so that high-level declarative knowledge, constraints, and non-monotonic inference fully participate in continuous optimisation and gradient-based learning. Instead of treating logic as a post-hoc filter on neural outputs, the method interprets each stable model as a configuration whose energy is the sum of neural energy factors attached to its atoms; inference finds the lowest-energy stable model, and learning lowers the energy of models that satisfy supervision relative to those that do not. The authors argue this removes the conceptual separation that has limited earlier neurosymbolic ASP systems in dynamic, partially observed settings. They supply a working implementation that injects neural energies into Clingo’s optimisation engine and demonstrate the pipeline on digit addition under sum-only supervision, compositional visual question answering on CLEVR, and multi-object tracking on MOT.

Core claim

ASPEn realises end-to-end neurosymbolic reasoning and learning by treating ASP stable models as preferred configurations inside a continuous energy landscape: logical rules restrict the feasible set while learnable energy factors attached to atoms rank those configurations, so that non-monotonic declarative semantics generate structured training signals for the neural substrate.

What carries the argument

Energised atoms of the form energise(rule_name([Args])) couple ASP atoms to neural energy functions; the energy of a stable model is the sum of its active energy factors, MAP inference is exact Clingo optimisation of that sum, and contrastive free-energy (F+ − F−) supplies the learning objective.

Load-bearing premise

The contrastive free-energy difference obtained by solving the same program once with and once without supervision constraints must yield usable gradients, and exact MAP optimisation over the grounded program must stay tractable for the combinatorial spaces that appear in real dynamic scenes.

What would settle it

Train the MNIST-addition or MOT pipeline with the published architecture; if digit-sum accuracy or HOTA/IDF1 collapses relative to the reported figures once the energy terms are replaced by uniform costs or the positive/negative free-energy contrast is ablated, the claim that ASP semantics meaningfully shape learning fails.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

Declarative background knowledge and non-monotonic defaults can be written once and then directly shape both inference and neural training without a separate loss-design step.
Perception pipelines for dynamic domains can maintain identity and spatial consistency by minimising an energy that already encodes hard logical constraints.
ASP becomes a practical substrate for end-to-end training rather than only a discrete post-processor of neural scores.
The same modular energy-atom interface can be reused across static VQA and multi-object tracking without redesigning the learning loop.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the free-energy contrast remains informative under heavier occlusion or longer temporal horizons, the method could support online belief revision in continuous video without re-engineering the neural backbone.
The energy-atom interface suggests a natural route for attaching other continuous modules (physics simulators, language embeddings) to the same ASP optimisation engine.
Benchmark suites that explicitly score how well learned energies respect non-monotonic defaults would become necessary once the method is adopted more widely.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

Clean ASP-EBM integration that works as a platform, but the full end-to-end free-energy claim is only shown on the toy case.

read the letter

The real move here is treating ASP stable models as configurations in an energy landscape: energise atoms, sum their EBM scores into E_theta(M|x), then do exact MAP via Clingo #minimize and train with the free-energy contrast L = F+ - F- (supervision constraints on vs off). That is a legitimate generalisation of the NeurASP / ASPMT / probabilistic-ASP line, and they ship a working Clingo+PyTorch loop plus three demos.

What they do well: the formal definitions are crisp, the MNIST addition experiment really is end-to-end under weak sum supervision and hits high digit/sum accuracy, and the MOT tables show that grounding and solving with hundreds of energy atoms stays cheap enough for real video. CLEVR and MOT at least prove the modular pipeline can be wired into existing detectors and trackers.

The soft spot is exactly the one the stress-test flags, and it is not minor. Only MNIST uses the ASP-derived free-energy gradients to train the network; CLEVR trains attribute heads with ordinary supervised hinge loss and uses ASP only at inference, while MOT freezes a pairwise contrastive EBM and lets ASP do assignment. So the headline claim of “joint optimisation o continuous latent space through explicit ASP-based declarative semantics” is demonstrated only on a 100-model space. Accuracies are modest (CLEVR 62.85 %, MOT HOTA 43), baselines and ablations are missing, and no code is released. That does not kill the framework, but it does mean the “robust o dynamic domains” rhetoric runs ahead of the evidence.

This is for people already building ASP-centric neurosymbolic systems who want a reusable energy substrate. It is not a breakthrough result, but it is a usable platform paper with honest formal scaffolding. I would send it to referees; the core construction is sound enough to improve under review. Worth a look if you are in the area; not urgent otherwise.

Referee Report

3 major / 4 minor

Summary. The paper proposes ASPEn, a neurosymbolic framework that couples Answer Set Programming with Energy-Based Models so that stable models become preferred configurations in a continuous energy landscape. Energised atoms (energise(...)) inject neural energy factors into an ASP program; inference is exact MAP via Clingo’s #minimize over the sum of active energies Eθ(M|x); learning uses a contrastive free-energy objective L = F+ − F− obtained by solving once with and once without supervision constraints. Three case studies are presented: weakly supervised MNIST digit addition (end-to-end), CLEVR visual question answering (attribute EBMs trained with supervised hinge loss, ASP used at inference), and MOT multi-object tracking (pairwise EBM trained offline, ASP used for constrained assignment). Runtime tables for MOT show that grounding and solving remain practical for hundreds of energy atoms per frame.

Significance. If the claimed joint optimisation truly works, ASPEn would give the ASP community a clean, modular route to end-to-end training that preserves non-monotonic semantics, defaults and abductive model selection—capabilities that probabilistic logic and neural-predicate approaches typically sacrifice. The practical Clingo/PyTorch implementation and the explicit energy-based semantics over stable models are genuine engineering contributions. The MNIST addition experiment already shows that weak supervision via ASP constraints can train a digit classifier to high accuracy. However, the stronger claim of “robust, end-to-end training for dynamic domains” is only partially substantiated by the current experiments, so the significance remains conditional on a clearer demonstration that ASP-derived free energies produce usable gradients on non-trivial combinatorial spaces.

major comments (3)

Sec. 3.2 defines the contrastive free-energy objective L = F+ − F− as the mechanism that lets non-monotonic ASP semantics participate in gradient-based training. Only the MNIST Addition experiment (Sec. 4.1) actually uses this objective; the hypothesis space is tiny (100 models). In CLEVR (Sec. 4.2) attribute EBMs are trained with a fully supervised margin hinge loss; ASP is used only for inference. In MOT (Sec. 4.3) the EBM is trained offline with pairwise contrastive loss and then frozen. Consequently the central claim of joint optimisation that “fully incorporates o non-monotonic inference” is not demonstrated on any non-trivial combinatorial space. Either an end-to-end free-energy training experiment on CLEVR or MOT, or an explicit ablation showing that the free-energy gradients remain informative when the number of stable models grows, is required.
CLEVR VQA accuracy is reported as 62.85 % (Fig. 3b) with no comparison to the neuro-symbolic ASP baselines already cited (Eiter et al. 2022, Yi et al. 2018) and no ablations that isolate the contribution of energy-based ranking versus pure ASP or pure neural attribute prediction. Without these numbers the claim that the energy formulation “successfully bridges visual perception and logical reasoning” cannot be evaluated.
MOT results (HOTA 43.26, MOTA 36.37, IDF1 49.89) are described as “comparable with trackers of similar character,” yet no concrete baseline numbers or public leaderboard references are supplied. Tables 1–2 report only inference-time statistics; they do not address training-time cost or gradient quality of the free-energy objective. A short comparison table against at least one standard tracking-by-detection method is needed to support the dynamic-domain claim.

minor comments (4)

Fig. 1 caption and surrounding text use both “energised” and “energized”; pick one spelling consistently.
Sec. 4.2: the functional-program-to-ASP translation is only illustrated by one example; a short formal sketch or reference to the translation algorithm would improve reproducibility.
Table 2 reports “Number of Energy Atoms per Frame 339 o 1058”; it would be useful to state whether these counts include only match atoms or also the auxiliary energy atoms generated by grounding.
The IoU threshold used as a hard constraint in MOT is listed as a free parameter but never given a concrete value or sensitivity analysis.

Circularity Check

0 steps flagged

No load-bearing circularity: ASP program and energy factors are independent inputs; contrastive free-energy loss and MAP inference do not reduce claimed results to fitted parameters by construction.

full rationale

The paper's central derivation (Sec. 3) defines an ASPEn program as the union of a classical ASP program (supplied by the modeller) with energised atoms whose energies come from independently parameterised neural factors f_θ. Stable-model energy is the additive sum of active factors; inference is exact MAP via Clingo #minimize; learning is ordinary contrastive free-energy L = F+ − F− obtained by solving once with and once without external supervision constraints C. None of these steps is definitional of the other: the ASP rules are not derived from data, the neural parameters are not obtained by renaming ASP atoms, and the reported accuracies (MNIST digit/sum, CLEVR attributes/VQA, MOT HOTA/MOTA/IDF1) are measured against external held-out labels. Self-citations (to the authors' prior ASPMT(QS) and visual-commonsense papers) appear only in motivation and related-work sections and are not invoked as uniqueness theorems or as the sole justification for any equation. Experiments on public benchmarks further confirm that the method is falsifiable outside its own fitted values. The only minor self-referential element is the reuse of the authors' earlier modelling style for space/motion, which does not force the energy-based claims. Hence circularity is negligible.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 2 invented entities

The framework rests on standard ASP stable-model semantics, the additive energy decomposition of EBMs, and the modelling choice that supervision can be expressed as logical constraints whose free-energy difference yields useful gradients. No new physical constants or free parameters are fitted to produce a claimed universal law; the free parameters are ordinary neural weights and a few hand-chosen thresholds (IoU, margin). The invented entities are the ‘energised atom’ construct and the ASPEn program itself.

free parameters (3)

neural energy-function weights θ
Learned by gradient descent on the contrastive free-energy loss; ordinary model parameters, not claimed as universal constants.
IoU threshold for hard spatial constraints (MOT)
Hand-chosen cutoff that forbids matches below a fixed IoU; affects the feasible set of stable models.
margin m in hinge loss (CLEVR attribute EBMs)
Margin used to push correct attribute energies below incorrect ones; standard but free hyper-parameter.

axioms (4)

standard math Stable-model semantics of ASP (Gelfond–Lifschitz reduct) correctly characterises preferred world models under non-monotonic assumptions.
Taken as given from the ASP literature (Sec. 2.1).
domain assumption Additive energy Eθ(M|x) = Σ fθ(x,a) over active energy atoms is a valid preference measure over stable models.
Core modelling choice of the paper (Sec. 3.1); inherits the usual EBM assumption that relative energies define preferences.
ad hoc to paper Contrastive free-energy difference F+ − F− obtained by solving with vs. without supervision constraints yields informative gradients for the energy networks.
Assumed without theoretical convergence guarantees (Sec. 3.2).
domain assumption Exact MAP inference via Clingo’s #minimize remains tractable for the grounded programs that arise in the target dynamic domains.
Supported only by the modest MOT runtime tables; not proved in general.

invented entities (2)

energised atom / energise(rule_name([Args])) predicate no independent evidence
purpose: Couples an ASP atom to a neural energy function so that logical atoms become continuous scores.
Syntactic and semantic extension introduced by the paper; no independent existence outside ASPEn.
ASPEn program Π ≡ Π_ASP ∪ Π_En no independent evidence
purpose: Defines the joint discrete-continuous object over which inference and learning are performed.
The central modelling construct of the framework.

pith-pipeline@v1.1.0-grok45 · 18475 in / 3091 out tokens · 26536 ms · 2026-07-10T12:30:43.203213+00:00 · methodology

0 comments

read the original abstract

We present a general neurosymbolic reasoning and learning methodology based on a modular integration of answer set programming with an energy based model substrate. Key contributions are: (1) supporting joint optimisation in the continuous latent space through explicit ASP-based declarative semantics fully incorporating background knowledge, constraints, non-monotonic inference; and (2) advancing recent works at the interface of answer sets, probabilistic logic, and answer set modulo theories by providing a generalised model and practical platform for ASP-centric robust, end-to-end training for applications in dynamic domains (e.g., involving perception and interaction). We provide a practical implementation, and demonstrate basic use and application (with MNIST), and evaluate with the visual question-answering benchmark Clevr and the multi-object tracking benchmark MOT.

Figures

Figures reproduced from arXiv: 2607.08136 by Jakob Suchan, Julius Monsen, Mehul Bhatt, Salim Baloch.

**Figure 1.** Figure 1: ASPEn: End-to-End Inference and Learning Architecture 3 ASP Energised! Formal Framework and Implementation ASPEn is a novel methodology integrating answer set programming and energybased models for declarative neurosymbolic reasoning and learning in a manner that is end-to-end, and designed for challenging dynamic domains. Building upon recent and emerging works in neurosymbolic ASP (Sec 2, 5), we offer a… view at source ↗

**Figure 2.** Figure 2: MNIST Addition: (a) Example from MNIST; (b) Digit and Sum accuracy. the energy-based formulation, these logits are interpreted as negative energies associated with each digit value. Energy based Inference. At the inference stage, we are searching for the most likely assignment of digit images and values, which translates to minimising the overall energy of the stable model, as described in Sec. 3.2. For th… view at source ↗

**Figure 3.** Figure 3: Clevr Dataset: (a) Example Scene from CLEVR Dataset; (b) Individual attribute accuracy and VQA accuracy on 15,000 CLEVR image-question pairs. the visual feature vector and four normalized bounding box coordinates. The network consists of two hidden layers with 256 units each, employing layer normalization and ReLU activations. The output layer produces two energy values corresponding to small and large. T… view at source ↗

**Figure 4.** Figure 4: Multi-Object Tracking (MOT): (a) Abductive inference for maintaining object tracks; (b) Tracking performance with standard MOT metrics [PITH_FULL_IMAGE:figures/full_fig_p015_4.png] view at source ↗

discussion (0)

Reference graph

Works this paper leans on

31 extracted references · 31 canonical work pages · 2 internal anchors

[1]

AI HLEG: High-level expert group on artificial intelligence: Ethical guide- lines for trustworthy ai (2019), https://www.aepd.es/sites/default/files/2019-12/ ai-ethics-guidelines.pdf

work page 2019
[2]

In: IJCAI 2013, Procee

Bartholomew, M., Lee, J.: Functional stable model semantics and answer set pro- gramming modulo theories. In: IJCAI 2013, Procee. of 23rd Intern. Joint Conf. ASPEn –Answer Set Programming Energised! 17 on Artificial Intelligence. pp. 718–724. IJCAI/AAAI (2013), http://www.aaai.org/ ocs/index.php/IJCAI/IJCAI13/paper/view/6894

work page 2013
[3]

In: Chetouani, M., Dignum, V., Lukowicz, P., Sierra, C

Bhatt, M., Suchan, J.: Artificial visual intelligence - perceptual commonsense for human-centred cognitive technologies. In: Chetouani, M., Dignum, V., Lukowicz, P., Sierra, C. (eds.) Human-Centered Artificial Intelligence - Advanced Lectures, 18th European Advanced Course on AI, ACAI 2021, Berlin, Germany, Octo- ber 11-15, 2021, extended and improved lec...

work page doi:10.1007/978-3-031-24349-3_12 2021
[4]

Brewka, G., Eiter, T., Truszczyński, M.: Answer set programming at a glance. Comm. ACM54(12), 92–103 (2011). https://doi.org/10.1145/2043174.2043195

work page doi:10.1145/2043174.2043195 2011
[5]

Dendorfer, P., Rezatofighi, S.H., Milan, A., Shi, J., Cremers, D., Reid, I.D., Roth, S., Schindler, K., Leal-Taixé, L.: CVPR19 tracking and detection challenge: How crowded can it get? CoRRabs/1906.04567(2019), http://arxiv.org/abs/1906. 04567

work page internal anchor Pith review Pith/arXiv arXiv 1906
[6]

Eiter, T., Hadl, J., Ruiz, N.H., Lange, L., Oetsch, J., Scheuvens, B., Strötgen, J.: Explainable zero-shot visual question answering via logic-based reasoning. In: H. Gilpin, L., Giunchiglia, E., Hitzler, P., van Krieken, E. (eds.) Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning. Proceedings of Machine Learning Res...

work page 2025
[7]

Theory and Practice of Logic Programming22(5), 739–754 (2022)

Eiter, T., Higuera, N., Oetsch, J., Pritz, M.: A neuro-symbolic asp pipeline for visual question answering. Theory and Practice of Logic Programming22(5), 739–754 (2022). https://doi.org/10.1017/S1471068422000229

work page doi:10.1017/s1471068422000229 2022
[8]

EU Commission: Proposal for a regulation of the european parliament and of the council laying down harmonised rules on artificial intelligence (artificial intelligence act) and amending certain union legislative acts (2021), https://eur-lex.europa.eu/ legal-content/EN/TXT/?uri=CELEX:52021PC0206

work page 2021
[9]

Garcez, A.d., Lamb, L.C.: Neurosymbolic ai: the 3rd wave. Artif. Intell. Rev. 56(11), 12387–12406 (Mar 2023). https://doi.org/10.1007/s10462-023-10448-w, https://doi.org/10.1007/s10462-023-10448-w

work page doi:10.1007/s10462-023-10448-w 2023
[10]

Morgan & Claypool Publishers (2012)

Gebser, M., Kaminski, R., Kaufmann, B., Schaub, T.: Answer Set Solving in Prac- tice. Morgan & Claypool Publishers (2012)

work page 2012
[11]

Clingo = ASP + Control: Preliminary Report

Gebser, M., Kaminski, R., Kaufmann, B., Schaub, T.: Clingo = ASP + control: Preliminary report. CoRRabs/1405.3694(2014)

work page internal anchor Pith review Pith/arXiv arXiv 2014
[12]

In: 2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA, July 21-26, 2017

Johnson, J., Hariharan, B., van der Maaten, L., Fei-Fei, L., Zitnick, C.L., Gir- shick, R.B.: CLEVR: A diagnostic dataset for compositional language and ele- mentary visual reasoning. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA, July 21-26, 2017. pp. 1988–

work page 2017
[13]

https://doi.org/10.1109/CVPR.2017.215, https://doi.org/10.1109/CVPR.2017.215

IEEE Computer Society (2017). https://doi.org/10.1109/CVPR.2017.215, https://doi.org/10.1109/CVPR.2017.215

work page doi:10.1109/cvpr.2017.215 2017
[14]

In: Bakir, G., Hofman, T., Schölkopf, B., Smola, A., Taskar, B

LeCun,Y.,Chopra,S.,Hadsell,R.,Ranzato,M.,Huang,F.J.:Atutorialonenergy- based learning. In: Bakir, G., Hofman, T., Schölkopf, B., Smola, A., Taskar, B. (eds.) Predicting Structured Data. pp. 192–241. MIT Press (2006)

work page 2006
[15]

lecun.com/exdb/mnist/

LeCun, Y., Cortes, C.: MNIST handwritten digit database (2010), http://yann. lecun.com/exdb/mnist/

work page 2010
[16]

Theory Pract

Lee, J., Talsania, S., Wang, Y.: Computing LPMLN using ASP and MLN solvers. Theory Pract. Log. Program.17(5-6), 942–960 (2017). https://doi.org/10.1017/ S1471068417000400

work page 2017
[17]

Springer (2019)

Lifschitz, V.: Answer Set Programming. Springer (2019). https://doi.org/10.1007/ 978-3-030-24658-7, https://doi.org/10.1007/978-3-030-24658-7 18 Suchan et al

work page doi:10.1007/978-3-030-24658-7 2019
[18]

In: Bengio, S., Wallach, H.M., Larochelle, H., Grauman, K., Cesa-Bianchi, N., Garnett, R

Manhaeve, R., Dumancic, S., Kimmig, A., Demeester, T., Raedt, L.D.: Deep- problog: Neural probabilistic logic programming. In: Bengio, S., Wallach, H.M., Larochelle, H., Grauman, K., Cesa-Bianchi, N., Garnett, R. (eds.) Advances in Neural Information Processing Systems 31: Annual Conference on Neural Infor- mation Processing Systems 2018, NeurIPS 2018, De...

work page 2018
[19]

Nordic Innovation: Nordic ethical ai expert group: Policy recommendations for ethical and responsible ai (2024), https://www.nordicinnovation.org/2024/ nordic-ethical-ai-expert-group

work page 2024
[20]

Theory and Practice of Logic Programming 25(4), 722–738 (2025)

Padalkar, P., Gupta, G.: Symbolic rule extraction from attention-guided sparse representations in vision transformers. Theory and Practice of Logic Programming 25(4), 722–738 (2025). https://doi.org/10.1017/S1471068425100318

work page doi:10.1017/s1471068425100318 2025
[21]

KI32(2-3), 101– 103 (2018)

Schaub, T., Woltran, S.: Special issue on answer set programming. KI32(2-3), 101– 103 (2018). https://doi.org/10.1007/s13218-018-0554-8, https://doi.org/10.1007/ s13218-018-0554-8

work page doi:10.1007/s13218-018-0554-8 2018
[22]

In: Kambhampati, S

Suchan, J., Bhatt, M.: Semantic question-answering with video and eye-tracking data: AI foundations for human visual perception driven cognitive film studies. In: Kambhampati, S. (ed.) Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI 2016, New York, NY, USA, 9-15 July

work page 2016
[23]

2633–2639

pp. 2633–2639. IJCAI/AAAI Press (2016), http://www.ijcai.org/Abstract/ 16/374

work page 2016
[24]

In: Ortiz, M., Wassermann, R., Schaub, T

Suchan, J., Bhatt, M., Monsen, J.: ASP Driven visual commonsense: A general framework for reasoning about embodied interaction in the wild. In: Ortiz, M., Wassermann, R., Schaub, T. (eds.) Proceedings of the 22nd International Confer- ence on Principles of Knowledge Representation and Reasoning, KR 2025, Mel- bourne, Australia, November 11-17, 2025 (2025)...

work page doi:10.24963/kr 2025
[25]

In: Proceedings of the 22nd International Conference on Principles of Knowledge Representation and Reasoning

Suchan, J., Bhatt, M., Monsen, J.: ASP-driven visual commonsense: a general framework for reasoning about embodied interaction in the wild. In: Proceedings of the 22nd International Conference on Principles of Knowledge Representation and Reasoning. KR ’25 (2025). https://doi.org/10.24963/kr.2025/61, https://doi. org/10.24963/kr.2025/61

work page doi:10.24963/kr.2025/61 2025
[26]

In: Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16,

Suchan, J., Bhatt, M., Varadarajan, S.: Out of sight but not out of mind: An answer set programming based online abduction framework for visual sensemaking in autonomous driving. In: Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16,

work page 2019
[27]

1879–1885

pp. 1879–1885. ijcai.org (2019). https://doi.org/10.24963/ijcai.2019/260

work page doi:10.24963/ijcai.2019/260 2019
[28]

Suchan, J., Bhatt, M., Varadarajan, S.: Commonsense visual sensemaking for au- tonomous driving - on generalised neurosymbolic online abduction integrating vi- sion and semantics. Artif. Intell.299, 103522 (2021). https://doi.org/10.1016/J. ARTINT.2021.103522, https://doi.org/10.1016/j.artint.2021.103522

work page doi:10.1016/j 2021
[29]

TPLP17(2), 205–225 (2017)

Wałęga, P.A., Schultz, C.P.L., Bhatt, M.: Non-monotonic spatial reasoning with answer set programming modulo theories. TPLP17(2), 205–225 (2017). https:// doi.org/10.1017/S1471068416000193,https://doi.org/10.1017/S1471068416000193

work page doi:10.1017/s1471068416000193 2017
[30]

In: Bessiere, C

Yang, Z., Ishay, A., Lee, J.: Neurasp: Embracing neural networks into answer set programming. In: Bessiere, C. (ed.) Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, IJCAI-20. pp. 1755–1762. International Joint Conferences on Artificial Intelligence Organization (7 2020). https://doi.org/ 10.24963/ijcai.2020/243, ...

work page doi:10.24963/ijcai.2020/243 2020
[31]

In: Pro- ceedings of the 32nd International Conference on Neural Information Processing Systems

Yi, K., Wu, J., Gan, C., Torralba, A., Kohli, P., Tenenbaum, J.B.: Neural-symbolic vqa: disentangling reasoning from vision and language understanding. In: Pro- ceedings of the 32nd International Conference on Neural Information Processing Systems. p. 1039–1050. NeurIPS’18, Curran Associates Inc., Red Hook, NY, USA (2018)

work page 2018

[1] [1]

AI HLEG: High-level expert group on artificial intelligence: Ethical guide- lines for trustworthy ai (2019), https://www.aepd.es/sites/default/files/2019-12/ ai-ethics-guidelines.pdf

work page 2019

[2] [2]

In: IJCAI 2013, Procee

Bartholomew, M., Lee, J.: Functional stable model semantics and answer set pro- gramming modulo theories. In: IJCAI 2013, Procee. of 23rd Intern. Joint Conf. ASPEn –Answer Set Programming Energised! 17 on Artificial Intelligence. pp. 718–724. IJCAI/AAAI (2013), http://www.aaai.org/ ocs/index.php/IJCAI/IJCAI13/paper/view/6894

work page 2013

[3] [3]

In: Chetouani, M., Dignum, V., Lukowicz, P., Sierra, C

Bhatt, M., Suchan, J.: Artificial visual intelligence - perceptual commonsense for human-centred cognitive technologies. In: Chetouani, M., Dignum, V., Lukowicz, P., Sierra, C. (eds.) Human-Centered Artificial Intelligence - Advanced Lectures, 18th European Advanced Course on AI, ACAI 2021, Berlin, Germany, Octo- ber 11-15, 2021, extended and improved lec...

work page doi:10.1007/978-3-031-24349-3_12 2021

[4] [4]

Brewka, G., Eiter, T., Truszczyński, M.: Answer set programming at a glance. Comm. ACM54(12), 92–103 (2011). https://doi.org/10.1145/2043174.2043195

work page doi:10.1145/2043174.2043195 2011

[5] [5]

Dendorfer, P., Rezatofighi, S.H., Milan, A., Shi, J., Cremers, D., Reid, I.D., Roth, S., Schindler, K., Leal-Taixé, L.: CVPR19 tracking and detection challenge: How crowded can it get? CoRRabs/1906.04567(2019), http://arxiv.org/abs/1906. 04567

work page internal anchor Pith review Pith/arXiv arXiv 1906

[6] [6]

Eiter, T., Hadl, J., Ruiz, N.H., Lange, L., Oetsch, J., Scheuvens, B., Strötgen, J.: Explainable zero-shot visual question answering via logic-based reasoning. In: H. Gilpin, L., Giunchiglia, E., Hitzler, P., van Krieken, E. (eds.) Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning. Proceedings of Machine Learning Res...

work page 2025

[7] [7]

Theory and Practice of Logic Programming22(5), 739–754 (2022)

Eiter, T., Higuera, N., Oetsch, J., Pritz, M.: A neuro-symbolic asp pipeline for visual question answering. Theory and Practice of Logic Programming22(5), 739–754 (2022). https://doi.org/10.1017/S1471068422000229

work page doi:10.1017/s1471068422000229 2022

[8] [8]

EU Commission: Proposal for a regulation of the european parliament and of the council laying down harmonised rules on artificial intelligence (artificial intelligence act) and amending certain union legislative acts (2021), https://eur-lex.europa.eu/ legal-content/EN/TXT/?uri=CELEX:52021PC0206

work page 2021

[9] [9]

Garcez, A.d., Lamb, L.C.: Neurosymbolic ai: the 3rd wave. Artif. Intell. Rev. 56(11), 12387–12406 (Mar 2023). https://doi.org/10.1007/s10462-023-10448-w, https://doi.org/10.1007/s10462-023-10448-w

work page doi:10.1007/s10462-023-10448-w 2023

[10] [10]

Morgan & Claypool Publishers (2012)

Gebser, M., Kaminski, R., Kaufmann, B., Schaub, T.: Answer Set Solving in Prac- tice. Morgan & Claypool Publishers (2012)

work page 2012

[11] [11]

Clingo = ASP + Control: Preliminary Report

Gebser, M., Kaminski, R., Kaufmann, B., Schaub, T.: Clingo = ASP + control: Preliminary report. CoRRabs/1405.3694(2014)

work page internal anchor Pith review Pith/arXiv arXiv 2014

[12] [12]

In: 2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA, July 21-26, 2017

Johnson, J., Hariharan, B., van der Maaten, L., Fei-Fei, L., Zitnick, C.L., Gir- shick, R.B.: CLEVR: A diagnostic dataset for compositional language and ele- mentary visual reasoning. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA, July 21-26, 2017. pp. 1988–

work page 2017

[13] [13]

https://doi.org/10.1109/CVPR.2017.215, https://doi.org/10.1109/CVPR.2017.215

IEEE Computer Society (2017). https://doi.org/10.1109/CVPR.2017.215, https://doi.org/10.1109/CVPR.2017.215

work page doi:10.1109/cvpr.2017.215 2017

[14] [14]

In: Bakir, G., Hofman, T., Schölkopf, B., Smola, A., Taskar, B

LeCun,Y.,Chopra,S.,Hadsell,R.,Ranzato,M.,Huang,F.J.:Atutorialonenergy- based learning. In: Bakir, G., Hofman, T., Schölkopf, B., Smola, A., Taskar, B. (eds.) Predicting Structured Data. pp. 192–241. MIT Press (2006)

work page 2006

[15] [15]

lecun.com/exdb/mnist/

LeCun, Y., Cortes, C.: MNIST handwritten digit database (2010), http://yann. lecun.com/exdb/mnist/

work page 2010

[16] [16]

Theory Pract

Lee, J., Talsania, S., Wang, Y.: Computing LPMLN using ASP and MLN solvers. Theory Pract. Log. Program.17(5-6), 942–960 (2017). https://doi.org/10.1017/ S1471068417000400

work page 2017

[17] [17]

Springer (2019)

Lifschitz, V.: Answer Set Programming. Springer (2019). https://doi.org/10.1007/ 978-3-030-24658-7, https://doi.org/10.1007/978-3-030-24658-7 18 Suchan et al

work page doi:10.1007/978-3-030-24658-7 2019

[18] [18]

In: Bengio, S., Wallach, H.M., Larochelle, H., Grauman, K., Cesa-Bianchi, N., Garnett, R

Manhaeve, R., Dumancic, S., Kimmig, A., Demeester, T., Raedt, L.D.: Deep- problog: Neural probabilistic logic programming. In: Bengio, S., Wallach, H.M., Larochelle, H., Grauman, K., Cesa-Bianchi, N., Garnett, R. (eds.) Advances in Neural Information Processing Systems 31: Annual Conference on Neural Infor- mation Processing Systems 2018, NeurIPS 2018, De...

work page 2018

[19] [19]

Nordic Innovation: Nordic ethical ai expert group: Policy recommendations for ethical and responsible ai (2024), https://www.nordicinnovation.org/2024/ nordic-ethical-ai-expert-group

work page 2024

[20] [20]

Theory and Practice of Logic Programming 25(4), 722–738 (2025)

Padalkar, P., Gupta, G.: Symbolic rule extraction from attention-guided sparse representations in vision transformers. Theory and Practice of Logic Programming 25(4), 722–738 (2025). https://doi.org/10.1017/S1471068425100318

work page doi:10.1017/s1471068425100318 2025

[21] [21]

KI32(2-3), 101– 103 (2018)

Schaub, T., Woltran, S.: Special issue on answer set programming. KI32(2-3), 101– 103 (2018). https://doi.org/10.1007/s13218-018-0554-8, https://doi.org/10.1007/ s13218-018-0554-8

work page doi:10.1007/s13218-018-0554-8 2018

[22] [22]

In: Kambhampati, S

Suchan, J., Bhatt, M.: Semantic question-answering with video and eye-tracking data: AI foundations for human visual perception driven cognitive film studies. In: Kambhampati, S. (ed.) Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI 2016, New York, NY, USA, 9-15 July

work page 2016

[23] [23]

2633–2639

pp. 2633–2639. IJCAI/AAAI Press (2016), http://www.ijcai.org/Abstract/ 16/374

work page 2016

[24] [24]

In: Ortiz, M., Wassermann, R., Schaub, T

Suchan, J., Bhatt, M., Monsen, J.: ASP Driven visual commonsense: A general framework for reasoning about embodied interaction in the wild. In: Ortiz, M., Wassermann, R., Schaub, T. (eds.) Proceedings of the 22nd International Confer- ence on Principles of Knowledge Representation and Reasoning, KR 2025, Mel- bourne, Australia, November 11-17, 2025 (2025)...

work page doi:10.24963/kr 2025

[25] [25]

In: Proceedings of the 22nd International Conference on Principles of Knowledge Representation and Reasoning

Suchan, J., Bhatt, M., Monsen, J.: ASP-driven visual commonsense: a general framework for reasoning about embodied interaction in the wild. In: Proceedings of the 22nd International Conference on Principles of Knowledge Representation and Reasoning. KR ’25 (2025). https://doi.org/10.24963/kr.2025/61, https://doi. org/10.24963/kr.2025/61

work page doi:10.24963/kr.2025/61 2025

[26] [26]

In: Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16,

Suchan, J., Bhatt, M., Varadarajan, S.: Out of sight but not out of mind: An answer set programming based online abduction framework for visual sensemaking in autonomous driving. In: Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16,

work page 2019

[27] [27]

1879–1885

pp. 1879–1885. ijcai.org (2019). https://doi.org/10.24963/ijcai.2019/260

work page doi:10.24963/ijcai.2019/260 2019

[28] [28]

Suchan, J., Bhatt, M., Varadarajan, S.: Commonsense visual sensemaking for au- tonomous driving - on generalised neurosymbolic online abduction integrating vi- sion and semantics. Artif. Intell.299, 103522 (2021). https://doi.org/10.1016/J. ARTINT.2021.103522, https://doi.org/10.1016/j.artint.2021.103522

work page doi:10.1016/j 2021

[29] [29]

TPLP17(2), 205–225 (2017)

Wałęga, P.A., Schultz, C.P.L., Bhatt, M.: Non-monotonic spatial reasoning with answer set programming modulo theories. TPLP17(2), 205–225 (2017). https:// doi.org/10.1017/S1471068416000193,https://doi.org/10.1017/S1471068416000193

work page doi:10.1017/s1471068416000193 2017

[30] [30]

In: Bessiere, C

Yang, Z., Ishay, A., Lee, J.: Neurasp: Embracing neural networks into answer set programming. In: Bessiere, C. (ed.) Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, IJCAI-20. pp. 1755–1762. International Joint Conferences on Artificial Intelligence Organization (7 2020). https://doi.org/ 10.24963/ijcai.2020/243, ...

work page doi:10.24963/ijcai.2020/243 2020

[31] [31]

In: Pro- ceedings of the 32nd International Conference on Neural Information Processing Systems

Yi, K., Wu, J., Gan, C., Torralba, A., Kohli, P., Tenenbaum, J.B.: Neural-symbolic vqa: disentangling reasoning from vision and language understanding. In: Pro- ceedings of the 32nd International Conference on Neural Information Processing Systems. p. 1039–1050. NeurIPS’18, Curran Associates Inc., Red Hook, NY, USA (2018)

work page 2018