Pith. sign in

REVIEW 3 major objections 2 minor 58 references

Zero-shot Compositional Action Recognition with Neural Logic Constraints

T0 review · 3 major / 2 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read LogicCAR claims logic constraints beat zero-shot action baselines, but the submitted text contains none of the claimed framework or experiments.

desk verdict The submission is a metadata/full-text mismatch: the abstract promises a ZS-CAR logic-constraint method, but the body is an unrelated CFD paper, so the central claim is unsupported and the paper cannot be reviewed as submitted. read the letter →

arxiv 2508.02320 v2 pith:X2XBUECZ submitted 2025-08-04 cs.CV

Gefan Ye , Lin Li , Kexin Li , Jun Xiao , Long Chen This is my paper
classification cs.CV
keywords zero-shotcompositionalactionrecognitionverb-objectcompositionfirst-orderlogicconstraintssemantichierarchyneural-symbolicreasoningSth-comdataset
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

The paper is trying to establish that zero-shot compositional action recognition improves when a neural model is constrained by two kinds of symbolic logic: one that restricts which verb-object compositions are plausible, and one that encodes a semantic hierarchy over action primitives. The authors argue that human-like symbolic reasoning fixes two weaknesses in current compositional models, namely spurious correlations between primitives and neglected semantic dependencies, and they propose a framework called LogicCAR that formalizes these constraints in first-order logic and embeds them into neural networks. They claim that LogicCAR outperforms existing baselines on the Sth-com dataset. Crucially, the full text supplied with this submission is a different paper about computational fluid dynamics in inhalers, so the abstract's claims are not accompanied by any of the described method, experiments, or comparisons.

What carries the argument

The central mechanism is the pair of symbolic constraints integrated into a neural network: Explicit Compositional Logic, which restricts which verb-object pairs are compositionally valid to reduce spurious correlations, and Hierarchical Primitive Logic, which models semantic dependencies among primitives to give the model a fine-to-coarse reasoning path. Both constraints are formalized in first-order logic and embedded into the neural architecture, which is intended to connect symbolic abstraction with learned video representations. The paper gives no equations or implementation details in the supplied text.

What would settle it

Run the described LogicCAR framework on the Sth-com zero-shot splits and compare it with a baseline that removes both logic constraints; if the unconstrained model matches or beats it, the central claim that explicit compositional and hierarchical logic constraints improve zero-shot compositional action recognition is refuted.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is that adding dual symbolic constraints to a zero-shot compositional action recognition model improves its ability to recognize unseen verb-object compositions. Explicit Compositional Logic models restrictions within compositions, while Hierarchical Primitive Logic captures semantic dependencies among different primitives and enables fine-to-coarse reasoning; both are formalized in first-order logic and embedded into a neural framework called LogicCAR. The authors state that extensive experiments on the Sth-com dataset show LogicCAR outperforming existing baselines. As submitted, the body text does not contain this framework or these experiments, so the core discovery exists only as the abstract's assertion.

Load-bearing premise

The load-bearing premise is that the hand-defined first-order logic rules and the assumed semantic hierarchy over verbs and objects match the actual structure of the Sth-com benchmark, so they act as helpful inductive bias rather than as an arbitrary restriction.

Editorial extensions

If this is right

  • If the claim is correct, adding compositional and hierarchical logic constraints would improve zero-shot generalization to unseen verb-object pairs in video action recognition.
  • The explicit compositional constraint should suppress spurious correlations between primitives that occur when certain verb-object pairs are never seen together during training.
  • The hierarchical primitive constraint should let the model reason from fine-grained primitive evidence to coarse action categories, potentially improving accuracy on semantically related unseen compositions.
  • The framework would support the broader idea that hand-specified symbolic structure can be injected into neural video models to improve compositional generalization.
  • If the Sth-com result holds, the same constraint recipe could be transferred to other compositional recognition benchmarks.

Reading between the lines

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

  • Because the supplied full text is an unrelated manuscript, the abstract's experimental claim is currently unfalsified in this submission; obtaining the actual LogicCAR paper and checking its ablations would be the immediate next step.
  • The hand-defined rule inventory and primitive hierarchy are a form of human bias: if the benchmark's latent structure does not match the rules, the constraints would hurt rather than help zero-shot accuracy.
  • A testable extension would be to compare LogicCAR against a baseline with randomly shuffled or partially corrupted logic rules; if the corrupted rules perform equally well, the claimed benefit would likely come from the neural backbone rather than the logic.
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

3 major / 2 minor

Summary. The submission's abstract proposes LogicCAR, a zero-shot compositional action recognition (ZS-CAR) framework that integrates two symbolic constraint families, Explicit Compositional Logic and Hierarchical Primitive Logic, formalized in first-order logic and embedded into neural network architectures, and claims that extensive experiments on the Sth-com dataset show that LogicCAR outperforms existing baselines. The provided full text, however, is a computational fluid dynamics paper on flash-boiling flow inside pressurized metered dose inhalers, with governing equations, turbulence modeling, grid-sensitivity analysis, and validation against inhaler experiments. None of the described LogicCAR framework, its logic formalization, its neural architecture, or any experiment on Sth-com appears in the body of the manuscript.

Significance. If the claimed results were substantiated, the contribution would be of genuine interest to the compositional zero-shot learning community: explicit logic constraints over compositional and hierarchical structure are a plausible inductive bias for improving generalization to unseen verb-object pairs, and Sth-com is a standard benchmark. However, as submitted, the manuscript contains none of the described method, formalization, or experiments, so the significance cannot be assessed. The paper offers no reproducible code, machine-checked proofs, or parameter-free derivations that would allow independent verification of the central claim.

major comments (3)
  1. [Full text, Sections 1-6 and Eqs. (1)-(15)] The entire body of the manuscript is a CFD study of flash-boiling in pressurized metered dose inhalers, covering device geometry, multiphase governing equations, turbulence modeling, grid sensitivity, and comparison with inhaler experiments. There is no definition of Explicit Compositional Logic, Hierarchical Primitive Logic, or any first-order-logic constraint, and no description of a neural architecture for action recognition. The abstract's central claim of a 'logic-driven ZS-CAR framework' is therefore entirely unsupported by the submitted text.
  2. [Abstract, 'Extensive experiments on the Sth-com dataset'] The abstract asserts that LogicCAR outperforms existing baseline methods on Sth-com, but the manuscript contains no dataset description, no evaluation protocol, no baseline comparisons, no result tables, and no figures reporting accuracy or other metrics on Sth-com. The only quantitative results in the paper, in Sections 4.2-4.6 and Figs. 10-21, validate a CFD model against inhaler measurements and are unrelated to compositional action recognition.
  3. [Method description (absent)] The claimed rule inventory and constraint weights, which would be the free parameters of the logic constraints, are not specified anywhere. Because neither the formalization of the constraints nor any ablation is present, the core claim that these constraints improve zero-shot generalization rather than act as tuned bias cannot be checked for circularity or overfitting. This omission is load-bearing: the abstract's purpose for the logic constraints is precisely to improve generalization, and without the formalization and experiments the claim is vacuous.
minor comments (2)
  1. [Header] The header lists arXiv:2508.02320 (cs.CV), but the body is clearly a physics paper (arXiv:2508.02316v1, physics.flu-dyn); the identifier/content mismatch should be corrected.
  2. [Abstract] The abbreviation 'Sth-com' is used without expansion; a definition (Something-Something) is needed in any revised version.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity detected: the supplied body text is an unrelated CFD paper, so the abstract's claim is unevidenced rather than circular.

full rationale

The abstract claims that LogicCAR, which 'integrates dual symbolic constraints' formalized in first-order logic, 'outperforms existing baseline methods' on the Sth-com dataset. However, the full text provided is a computational fluid dynamics study of flash-boiling in pressurized metered dose inhalers, with no description of LogicCAR, its logic constraints, neural architecture, or Sth-com experiments. Because there is no derivation chain in the supplied document, there is no equation or fitted parameter that can be shown to reduce to its own inputs. The hard rule for this pass requires exhibiting a specific reduction (e.g., Eq. X = Eq. Y by construction, or a fitted parameter renamed as a prediction). No such reduction can be exhibited here. The mismatch between the abstract and the body is a completeness and reproducibility problem, not a circularity problem. Therefore the honest finding is no significant circularity, with score 0.

Assumptions & free parameters 1 free parameters · 3 assumptions · 2 invented entities

Because the full text does not match the abstract, this ledger reflects only what the abstract asserts. The two proposed constraint modules are the only new artifacts introduced; their rule definitions and weighting are undisclosed degrees of freedom treated here as free parameters. The standard zero-shot factorization assumptions are inherited from the problem definition, and the benchmark's validity is assumed. No additional entities, forces, or physical postulates are introduced.

free parameters (1)
  • Logic rule inventory and constraint weights = not reported
    The compositional and hierarchical logic rules must be chosen by hand, including which rules to encode, how many, and how strongly they are enforced relative to the data loss. The abstract discloses none of these choices, so any benchmark gain could in principle come from a tuned rule set rather than from the logic principle itself.
assumptions (3)
  • domain assumption Actions factor into independent verb and object primitives learned from seen compositions
    The abstract's first sentence defines ZS-CAR as recognizing unseen verb-object compositions from knowledge of verb and object primitives learned during training; this factorization is the inductive basis of everything that follows.
  • domain assumption First-order logic constraints can be embedded into neural networks and optimized by gradient descent
    The abstract states the constraints are formalized in first-order logic and embedded into neural network architectures, but no differentiable encoding is described; the viability of the embedding is assumed.
  • domain assumption Sth-com is a valid benchmark for measuring compositional generalization of actions
    The only empirical evidence claimed is performance on the Sth-com dataset; the abstract assumes gains on this benchmark measure the two proposed constraints rather than benchmark-specific artifacts.
invented entities (2)
  • Explicit Compositional Logic constraint
    purpose: Models restrictions within verb-object compositions to reduce spurious correlations between primitives during zero-shot recognition
    A proposed model component. The only evidence for its benefit would be the claimed Sth-com results, which are absent from the provided text, so no falsifiable handle exists outside this paper.
  • Hierarchical Primitive Logic constraint
    purpose: Models semantic dependencies among primitives to support fine-to-coarse reasoning during zero-shot recognition
    A proposed model component. As with the compositional constraint, the only evidence for its benefit would be the claimed Sth-com results, which are absent from the provided text.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Zero-shot Compositional Action Recognition with Neural Logic Constraints." pith.science (2026). https://pith.science/paper/X2XBUECZ

@misc{pith2026250802320,
  author       = {Pith},
  title        = {Pith review of: Zero-shot Compositional Action Recognition with Neural Logic Constraints},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/X2XBUECZ}},
  note         = {Machine review of arXiv:2508.02320}
}
read the original abstract

Zero-shot compositional action recognition (ZS-CAR) aims to identify unseen verb-object compositions in the videos by exploiting the learned knowledge of verb and object primitives during training. Despite compositional learning's progress in ZS-CAR, two critical challenges persist: 1) Missing compositional structure constraint, leading to spurious correlations between primitives; 2) Neglecting semantic hierarchy constraint, leading to semantic ambiguity and impairing the training process. In this paper, we argue that human-like symbolic reasoning offers a principled solution to these challenges by explicitly modeling compositional and hierarchical structured abstraction. To this end, we propose a logic-driven ZS-CAR framework LogicCAR that integrates dual symbolic constraints: Explicit Compositional Logic and Hierarchical Primitive Logic. Specifically, the former models the restrictions within the compositions, enhancing the compositional reasoning ability of our model. The latter investigates the semantical dependencies among different primitives, empowering the models with fine-to-coarse reasoning capacity. By formalizing these constraints in first-order logic and embedding them into neural network architectures, LogicCAR systematically bridges the gap between symbolic abstraction and existing models. Extensive experiments on the Sth-com dataset demonstrate that our LogicCAR outperforms existing baseline methods, proving the effectiveness of our logic-driven constraints.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

58 extracted references · 51 canonical work pages

  1. [1]

    Finlay, The mechanics of inhaled pharmaceutical aerosols, Academic Press, 2019

    W. Finlay, The mechanics of inhaled pharmaceutical aerosols, Academic Press, 2019

    work page 2019

  2. [2]

    de Boer, K

    A. de Boer, K. Thalberg, Devices and formulations: General introduc- tion and wet aerosol delivery systems, in: S. Kassinos, P. B¨ ackman, J. Conway, A. Hickey (Eds.), Inhaled Medicines, Academic Press, 2021, Ch. 3, pp. 35–63. doi:10.1016/B978-0-12-814974-4.00011-0

    work page doi:10.1016/b978-0-12-814974-4.00011-0 2021

  3. [3]

    de Boer, K

    A. de Boer, K. Thalberg, Metered dose inhalers MDIs, in: S. Kassinos, P. B¨ ackman, J. Conway, A. Hickey (Eds.), Inhaled Medicines, Academic Press, 2021, Ch. 4, pp. 65–97. doi:10.1016/B978-0-12-814974-4.00001-8

    work page doi:10.1016/b978-0-12-814974-4.00001-8 2021

  4. [4]

    H. D. Smyth, G. Brace, T. Barbour, J. Gallion, J. Grove, A. J. Hickey, Spray pattern analysis for metered dose inhalers: Effect of actuator design, Pharm. Res. 23 (2006) 1591–1596. doi:10.1007/s11095-006-0280- z

    work page doi:10.1007/s11095-006-0280- 2006

  5. [5]

    B. Zhu, D. Traini, P. Young, Aerosol particle generation from solution- based pressurized metered dose inhalers: a technical overview of pa- rameters that influence respiratory deposition, Pharm. Dev. Technol. 20 (2015) 897–910. doi:10.3109/10837450.2014.959176

    arXiv 2015

  6. [6]

    Legh-Land, A

    V. Legh-Land, A. Haddrell, D. Lewis, D. Murnane, J. Reid, Water up- take by evaporating pMDI aerosol prior to inhalation affects both re- gional and total deposition in the respiratory system, Pharmaceutics 13 (2021). doi:10.3390/pharmaceutics13070941

    work page doi:10.3390/pharmaceutics13070941 2021

  7. [7]

    Sheth, M

    P. Sheth, M. Grimes, S. Stein, P. Myrdal, Impact of droplet evap- oration rate on resulting in vitro performance parameters of pres- surized metered dose inhalers, Int. J. Pharm. 528 (2017) 360–371. doi:10.1016/j.ijpharm.2017.06.014. 44

    work page doi:10.1016/j.ijpharm.2017.06.014 2017

  8. [8]

    Stein, P

    S. Stein, P. Myrdal, The relative influence of atomization and evap- oration on metered dose inhaler drug delivery efficiency, Aerosol Sci. Technol. 40 (2006) 335–347. doi:10.1080/02786820600612268

    work page doi:10.1080/02786820600612268 2006

Show all 58 references
  1. [9]

    Yousefi, K

    M. Yousefi, K. Inthavong, J. Tu, Effect of pressurized metered dose inhaler spray characteristics and particle size distribution on drug de- livery efficiency, J. Aerosol Med. Pulm. Drug Deliv. 30 (2017) 359–372. doi:10.1089/jamp.2016.1299

    2017

  2. [10]

    Kakade, H

    P. Kakade, H. Versteeg, G. Hargrave, P. Genova, R. Williams III, D. Deaton, Design optimization of a novel pMDI actuator for systemic drug delivery, J. Aerosol Med. 20 (2007) 460–474. doi:10.1089/jam.2007.0595

    2007

  3. [11]

    Clark, Metered atomisation for respiratory drug delivery, Ph.D

    A. Clark, Metered atomisation for respiratory drug delivery, Ph.D. the- sis, Loughborough University (1991)

    1991

  4. [12]

    Chizari, H

    H. Chizari, H. Versteeg, W. Malalasekera, B. Gavtash, A novel mathe- matical model to predict the particle size distribution of pMDI aerosols, in: Proc. of DDL 2023, 2023

    2023

  5. [13]

    Fletcher, Factors affecting the atomization of saturated liquids, Ph.D

    G. Fletcher, Factors affecting the atomization of saturated liquids, Ph.D. thesis, Loughborough University (1975)

    1975

  6. [14]

    Dunbar, J

    C. Dunbar, J. Miller, Theoretical investigation of the spray from a pressurized metered-dose inhaler, Atomization Sprays 7 (1997) 417–436. doi:10.1615/AtomizSpr.v7.i4.60

    1997 doi

  7. [15]

    D. Ju, J. Shrimpton, A. Hearn, The effect of reduction of propellant mass fraction on the injection profile of metered dose inhalers, Int. J. Pharm. 391 (2010) 221–229. doi:10.1016/j.ijpharm.2010.03.003

    2010 doi

  8. [16]

    Gavtash, H

    B. Gavtash, H. Versteeg, G. Hargrave, B. Myatt, D. Lewis, T. Church, G. Brambilla, Transient flashing propellant flow models to predict internal flow characteristics, spray velocity, and aerosol droplet size of a pMDI, Aerosol Sci. Technol. 51 (2017) 564–575. doi:10.1080/02786...

    2017

  9. [17]

    Gavtash, H

    B. Gavtash, H. Versteeg, G. Hargrave, B. Myatt, D. Lewis, T. Church, B. G., A model of transient internal flow and at- omization of propellant/ethanol mixtures in pressurized metered 45 dose inhalers (pMDI), Aerosol Sci. Technol. 52 (2018) 494–504. doi:10.1080/02786826.2018.1433814

    2018

  10. [18]

    Alatrash, E

    A. Alatrash, E. Matida, Characterization of medication velocity and size distribution from pressurized metered-dose inhalers by phase doppler anemometry, J. Aerosol Med. Pulm. Drug Deliv. 29 (2016) 501–513. doi:10.1089/jamp.2015.1264

    2016

  11. [19]

    Crosland, M

    B. Crosland, M. Johnson, A. Matida, Edgar, Characterization of the spray velocities from a pressurized metered-dose inhaler, J. Aerosol Med. Pulm. Drug Deliv. 22 (2009) 85–98. doi:10.1089/jamp.2008.0687

    2009

  12. [20]

    Jahed, J

    M. Jahed, J. Kozinski, L. Pakzad, Mucus, airway and plume temper- ature effects on pMDI-drug delivery in a mouth-throat airway: Ex- perimental and numerical studies, Journal Aerosol Sci. 181 (2024). doi:10.1016/j.jaerosci.2024.106436

    2024

  13. [21]

    Kotak, C

    R. Kotak, C. Pandya, A. Pandya, A. Rajput, B. Thakur, Determina- tion of spray pattern and plume geometry of combined budesonide and formoterol fumarate pressurized metered dose inhalation aerosol, Int. J. Pharm. Sci. Res. 12 (2020) 614–620. doi:10.25004/IJPSDR.2020.120605

    2020

  14. [22]

    Moraga-Espinoza, E

    D. Moraga-Espinoza, E. Eshaghian, H. Smyth, Mass median plume angle: A novel approach to characterize plume geome- try in solution based pMDIs, Int. J. Pharm. 543 (2018) 376–385. doi:10.1016/j.ijpharm.2018.04.008

    2018 doi

  15. [23]

    Mason-Smith, D

    N. Mason-Smith, D. Duke, A. Kastengren, D. Traini, P. Young, Y. Chen, D. Lewis, D. Edgington-Mitchell, D. Honnery, Revealing pMDI spray initial conditions: flashing, atomisation and the effect of ethanol., Pharm. Res. 34 (2017) 718–729. doi:10.1007/s11095-017-2098-2

    2017 doi

  16. [24]

    McKiernan, Inhaler spray investigation using high-speed phase- contrast x-ray and schlieren imaging, Pharm

    A. McKiernan, Inhaler spray investigation using high-speed phase- contrast x-ray and schlieren imaging, Pharm. Res. 36 (2019). doi:10.1007/s11095-019-2657-9

    2019 doi

  17. [25]

    Gavtash, S

    B. Gavtash, S. Stein, H. Versteeg, B. Myatt, Application of theoreti- cal models to predict expansion chamber pressure and temperature of hfa134aethanol mixture, during a pMDI actuation, in: Proc. of DDL 30, 2019. 46

    2019

  18. [26]

    B. J. Myatt, H. K. Versteeg, G. K. Hargrave, E. J. Long, B. Gav- tash, D. A. Lewis, T. Church, G. Brambilla, Unlocking further understanding of the atomization mechanism of a pressurized me- tered dose inhaler, Aerosol Sci. Technol. 56 (2022) 1022–1032. doi:10.1080/02786826.20...

    2022

  19. [27]

    Myatt, D

    B. Myatt, D. Lewis, T. Church, G. Brambilla, G. Hargrave, H. Versteeg, L. E., B. Gavtash, Pda analysis of hfa/ethanol pMDI aerosols: An im- proved test protocol and new findings, in: Proceedings of the 13th Trien- nial International Conference on Liquid Atomization and Spray S...

    2015

  20. [28]

    Myatt, R

    B. Myatt, R. Newton, T. Church, G. Brambilla, G. Hargrave, H. Ver- steeg, B. Gavtash, L. E., Pda and high speed image analysis of hfa/ethanol pMDI aerosols, in: Proc. of DDL 26, 2015

    2015

  21. [29]

    Gavtash, Cfd simulation of pressurised metered-does inhaler (pMDI), Ph.D

    B. Gavtash, Cfd simulation of pressurised metered-does inhaler (pMDI), Ph.D. thesis, Loughborough University (2016)

    2016

  22. [30]

    J. Xi., T. Mohamed, X. Si, Two-way coupling and kolmogorov scales on inhaler spray plume evolutions from ventolin, proair, and qvar, Math. Biosci. Eng. 19 (2022) 10915–10940. doi:10.3934/mbe.2022510

    2022 doi

  23. [31]

    Oliveira, A

    R. Oliveira, A. Ferreira, S. Teixeira, J. Teixeira, H. Cabral-Marques, A CFD study of a pMDI plume spray, in: Transactions on Engineering Technologies, 2014

    2014

  24. [32]

    Dastoorian, L

    F. Dastoorian, L. Pakzad, J. Kozinski, E. Behzadfar, A CFD inves- tigation on the aerosol drug delivery in the mouth–throat airway us- ing a pressurized metered-dose inhaler device, Processes 10 (2022). doi:10.3390/pr10071230

    2022 doi

  25. [33]

    Ahookhosh, M

    K. Ahookhosh, M. Saidi, M. Mohammadpourfard, H. Aminfar, H. Hamishehkar, A. Farnoud, O. Schmid, Flow structure and parti- cle deposition analyses for optimization of a pressurized metered dose inhaler (pMDI) in a model of tracheobronchial airway, Eur. J. Pharm. Sci. 164 (2021)...

    2021

  26. [34]

    R. L. Walenga, G. Tian, P. Longest, Development of characteristic up- per tracheobronchial airway models for testing pharmaceutical aerosol delivery, J. Biomech. Eng. 135 (2013). doi:10.1115/1.4024630. 47

    2013 doi

  27. [35]

    H. G. Weller, G. Tabor, H. Jasak, C. Fureby, A tensorial approach to computational continuum mechanics using object-oriented techniques, Comput. Phys. 12 (1998) 620–631. doi:10.1063/1.168744

    1998 doi

  28. [36]

    B. D. Nichols, C. W. Hirt, Methods for calculating multi-dimensional, transient free surface flows past bodies, in: Proc. of 1st Int. Conf. on Num. Ship Hydrodyn, 1975

    1975

  29. [37]

    Tillner-Roth, H

    R. Tillner-Roth, H. D. Baehr, An international standard formulation of the thermodynamic properties of 1,1,1,2-tetrafluoroethane (hfc-134a) covering temperatures from 170 k to 455 k at pressures up to 70 mpa, J. Phys. Chem. Ref. Data 23 (1994) 657–729. doi:10.1063/1.555958

    1994 doi

  30. [38]

    A. K. Singhal, M. M. Athavale, H. Li, Y. Jiang, Mathematical basis and validation of the full cavitation model, J. Fluids Eng. 124 (2002) 617–624. doi:10.1115/1.1486223

    2002 doi

  31. [39]

    F. R. Menter, Two-equation eddy-viscosity turbulence models for engineering applications, AIAA J. 32 (1994) 1598–1605. doi:doi.org/10.2514/3.12149

    1994 doi

  32. [40]

    W. M. Kays, Turbulent prandtl number–where are we?, J. Heat Transf. 116 (1994) 284––295. doi:doi:10.1115/1.2911398

    1994 doi

  33. [41]

    X. Liao, D. Lucas, Computational modelling of flash boiling flows: A literature survey, Int. J. Heat Mass Transf. 111 (2017) 246–265. doi:10.1016/j.ijheatmasstransfer.2017.03.121

    2017 doi

  34. [42]

    I. K. Karathanassis, P. Koukouvinis, M. Gavaises, Compara- tive evaluation of phase-change mechanisms for the prediction of flashing flows, Int. J. Multiph. Flow 95 (2017) 257–270. doi:10.1016/j.ijmultiphaseflow.2017.06.006

    2017 doi

  35. [43]

    P. J. Zwart, A. G. Gerber, T. Belamri, A two-phase flow model for predicting cavitation dynamics, in: Proc. of 5th Int. Conf. on Multiph. Flow, 2004

    2004

  36. [44]

    B. C. E., Cavitation and bubble dynamics, Oxford University Press, 1995. 48

    1995

  37. [45]

    Theofanous, L

    T. Theofanous, L. Biasi, H. S. Isbin, H. Fauske, A theoretical study on bubble growth in constant and time-dependent pressure fields, Chem. Eng. Sci. 24 (1969) 885–897. doi:10.1016/0009-2509(69)85008-6

    1969 doi

  38. [46]

    Narayanan, Numerical simulation of flashing using a pressure- based compressible multiphase approach and a thermody- namic cavitation model, Int

    C. Narayanan, Numerical simulation of flashing using a pressure- based compressible multiphase approach and a thermody- namic cavitation model, Int. J. Multiph. Flow 135 (2021). doi:10.1016/j.ijmultiphaseflow.2020.103511

    2021

  39. [47]

    Weixing, J

    Y. Weixing, J. Sauer, G. H. Schnerr, Modeling and computation of un- steady cavitation flows in injection nozzles, M´ ec. Ind. 2 (2001) 383–394. doi:10.1016/S1296-2139(01)01120-4

    2001 doi

  40. [48]

    H. Xu, T. Wang, Z. Che, Flow in fuel nozzles under cavitation and flash-boiling conditions, AIP Adv. 12 (2022). doi:10.1016/0009- 2509(69)85008-6

    2022 doi

  41. [49]

    Zhang, X

    Y. Zhang, X. W. Luo, B. Ji, S. H. Liu, Y. L. Wu, H. Y. Xu, Computa- tional modelling of flash boiling flows: A literature survey, Chin. Phys. Lett. 27 (2010). doi:10.1088/0256-307X/27/1/016401

    2010 doi

  42. [50]

    X. Liao, D. Lucas, 3d cfd simulation of flashing flows in a converging-diverging nozzle, Nucl. Eng. Des. 292 (2015) 149–163. doi:10.1016/j.nucengdes.2015.06.015

    2015 doi

  43. [51]

    J. P. Janet, X. Liao, D. Lucas, Heterogeneous nucleation in cfd simu- lation of flashing flows in converging–diverging nozzles, Int. J. Multiph. Flow 74 (2015) 106–117. doi:10.1016/j.ijmultiphaseflow.2015.04.005

    2015 doi

  44. [52]

    Downar-Zapolski, Z

    P. Downar-Zapolski, Z. Bilicki, L. Bolle, J. Franco, The non-equilibrium relaxation model for one-dimensional flashing liquid flow, Int. J. Mul- tiph. Flow 3 (1996) 473–483. doi:10.1016/0301-9322(95)00078-X

    1996 doi

  45. [53]

    Lyras, S

    K. Lyras, S. Dembele, E. Vyazmina, S. Jallais, J. Wen, Numerical simu- lation of flash-boiling through sharp-edged orifices, Int. J. Comp. Meth. and Exp. Meas. 1 (2018) 176–185. doi:10.2495/CMEM-V6-N1-176-185

    2018 doi

  46. [54]

    R. F. Kunz, D. A. Boger, D. R. Stinebring, T. S. Chyczewski, L. J. W., H. J. Gibeling, S. Venkateswaran, T. Govindan, A precondi- tioned implicit method for two-phase flows with application to cavi- tation prediction, Comp. Fluids 29 (2000) 849–875. doi:10.1016/S0045- 7930(99)...

    2000 doi

  47. [55]

    C. L. Merkle, J. Feng, P. E. O. Buelow, Computational modeling of the dynamics of sheet cavitation, in: Proc. of 3rd Int. Symp. on Cavitation, 1998

    1998

  48. [56]

    G. H. Schnerr, J. Sauer, Physical and numerical modeling of unsteady cavitation dynamics, in: Proc. of 4th Int. Conf. on Multiph. Flow, 2001

    2001

  49. [57]

    Van Leer, Towards the ultimate conservative difference scheme

    B. Van Leer, Towards the ultimate conservative difference scheme. iv. a new approach to numerical convection, J. Comp. Physics 23 (1977) 276–299. doi:10.1016/0021-9991(77)90095-X

    1977 doi

  50. [58]

    D. Duke, D. Nguyen, H. Scott, A. Kusangaya, A. Kastengren, D. Schmidt, D. Honnery, Combining modelling with advanced exper- imental techniques to optimize formulation and nozzle design for low gwp pMDIs, in: Proceedings of Respiratory Drug Delivery 2021, 2021. 50

    2021

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.