Pith. sign in

REVIEW 4 major objections 4 minor 38 references

Spike Agreement Dependent Plasticity: A scalable Bio-Inspired learning paradigm for Spiking Neural Networks

T0 review · 4 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read The paper claims that SADP, a synaptic rule that learns from agreement between spike trains rather than precise spike pairing, beats classical STDP on MNIST and Fashion-MNIST while running in linear time.

desk verdict The abstract describes a genuinely new learning rule, but the full text is unreadable here and the accuracy claim hinges on an unverified kernel-independence assumption. read the letter →

arxiv 2508.16216 v1 pith:JLB7LWZ2 submitted 2025-08-22 cs.NE cs.LG

Saptarshi Bej , Muhammed Sahad E , Gouri Lakshmi , Harshit Kumar , Pritam Kar , Bikas C Das This is my paper · ORCID
classification cs.NEcs.LG
keywords spikingneuralnetworkssynapticplasticitySTDPCohen'skappaagreement-basedlearningneuromorphichardwarememtransistorMNIST
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 introduces Spike Agreement Dependent Plasticity (SADP), a learning rule for spiking neural networks that adjusts a synapse according to how much the pre- and post-synaptic spike trains agree, measured by a population-level correlation such as Cohen's kappa. This replaces classical STDP's reliance on the exact order and timing of individual spike pairs. The paper claims that the SADP update is linear-time, can be implemented with bitwise logic, and—when the agreement kernel is a spline fitted to measured iontronic organic memtransistor devices—is more accurate and faster than STDP on MNIST and Fashion-MNIST. If the claim holds, it implies that precise spike-pair timing is not required for effective biologically inspired learning, and that measured physical-device responses can serve directly as learning kernels.

What carries the argument

The central object is Cohen's kappa, a chance-corrected agreement coefficient that measures the population-level correspondence between pre- and post-synaptic spike trains; it replaces the temporally precise pairwise spike windows of STDP. The weight change for a given agreement level is set by a spline kernel fitted to the authors' measured iontronic organic memtransistor devices—transistor-like devices whose conductance acts as a tunable synaptic weight. This combination lets the learning rule be expressed as simple counts of agreement, which is what yields linear-time updates and bitwise-logic hardware implementation.

What would settle it

Train the same SADP network on MNIST with the device-derived spline kernel replaced by a shuffled or randomly generated spline of the same smoothness, keeping all other settings fixed. If accuracy does not drop, the device-specific kernel shape is not doing the causal work, and the reported advantage over STDP would be an artifact of kernel choice rather than a property of agreement-based learning.

Watch

Extended reading notes

Core claim

In the paper's own terms, SADP generalizes STDP by replacing pairwise temporal updates with population-level correlation metrics, specifically Cohen's kappa, so that a synapse changes when the two spike trains agree beyond what chance would predict. The update is computed from co-occurrence counts, giving linear-time complexity and a bitwise implementation that suits digital neuromorphic hardware. The empirical claim is that SADP, especially with spline-based kernels derived from the authors' experimental iontronic organic memtransistor device data, outperforms classical STDP in both accuracy and runtime on MNIST and Fashion-MNIST, bridging biological plausibility and computational scalabili

Load-bearing premise

The claimed gain over classical timing-based learning rests on the curves that set the weight change coming from the physical device measurements as a general prior, not on those curves being chosen because they happened to score well on the same MNIST and Fashion-MNIST tests used to report the improvement.

Editorial extensions

If this is right

  • Spiking networks could learn without circuits that measure precise millisecond timing; a count of spike-train agreement would suffice.
  • The learning rule is linear-time and bitwise-friendly, so it is a candidate for on-chip learning in neuromorphic systems where STDP's pairwise timing computations are expensive.
  • Device-derived spline kernels mean a physical synapse's measured behavior can be inserted directly into the learning rule, shortening the path from device measurements to trained networks.
  • If SADP's advantage holds, the standard STDP assumption that order and timing of individual spike pairs are the essential learning signal would need to be revised toward population-level statistics.

Reading between the lines

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

  • A direct test that follows from the rule's design is to run SADP on event-based sensor streams with artificially jittered spike times; an agreement-based update should degrade less than STDP as jitter grows, since timing precision is not part of the update.
  • Because the update is just a bitwise count, SADP could be fused with in-memory bitwise computing architectures, although the paper does not build or benchmark such a chip.
  • The Cohen's kappa formulation invites an information-theoretic reading of synaptic learning—treating the two spike trains as raters and learning as maximizing chance-corrected agreement—which the paper leaves unexplored.
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 introduces Spike Agreement Dependent Plasticity (SADP), a synaptic learning rule for spiking neural networks that replaces pairwise spike-timing plasticity with a population-level agreement measure (Cohen's kappa). The abstract claims that SADP generalizes classical STDP, has linear-time complexity, supports efficient bitwise hardware implementation, and—especially with spline-based kernels derived from the authors' experimental iontronic organic memtransistor device data—outperforms STDP on MNIST and Fashion-MNIST in both accuracy and runtime. The submitted manuscript, however, contains only an abstract followed by garbled and unreadable text; no equations, algorithm pseudocode, experimental protocol, numerical results, or code are accessible.

Significance. The conceptual idea—basing synaptic updates on spike-train agreement rather than precise spike-pair timing—is interesting and potentially relevant for neuromorphic hardware and biologically plausible learning. If the claimed accuracy and runtime improvements over STDP were substantiated with a full derivation, controlled experiments, and error bars, this could be a useful contribution. The paper does not currently ship machine-checked proofs, reproducible code, or detailed experimental tables; its main strength is that the headline claim is falsifiable, but the evidence provided is not sufficient to evaluate it.

major comments (4)
  1. [Full text (all sections after Abstract)] The body of the manuscript is unreadable: after the abstract, the text is a sequence of nonsensical characters with no equations, figures, tables, or coherent sentences. The central derivation of SADP, the definition of Cohen's kappa in the update rule, and the claimed reduction to linear-time complexity cannot be checked. This is not a minor presentation issue; it is a load-bearing absence of all technical content.
  2. [Abstract, results sentence] The claim that SADP 'outperforms classical STDP in both accuracy and runtime' is stated without any supporting numbers. There is no experimental setup, dataset split description, network architecture, spike encoding scheme, hyperparameter list, number of trials, error bars, or description of the STDP baseline. As presented, the accuracy and runtime claims are unsupported.
  3. [Abstract, spline kernels] The spline-based kernels are described as 'derived from our experimental iontronic organic memtransistor device data,' but the manuscript does not state whether the kernel parameters were fit independently of MNIST/Fashion-MNIST or selected using validation performance on those benchmarks. If the latter, the reported improvement over STDP would be a fitting artifact rather than a general property of SADP. The authors must clarify the kernel fitting procedure and provide sensitivity analyses over kernel choices not tuned to the target benchmarks.
  4. [Abstract, complexity and runtime claims] The 'linear-time complexity' claim is ambiguous: linear in the number of synapses, spikes, timesteps, or samples? No algorithm is provided. The runtime superiority claim also requires wall-clock measurements on identical hardware, along with code and hyperparameter settings, to be meaningful. Without these details, the scalability argument cannot be assessed.
minor comments (4)
  1. [Abstract] Cohen's kappa is not defined or cited. Please provide the formula and explain how it is computed from pre- and post-synaptic spike trains.
  2. [Abstract] The phrase 'bio-Inspired' has inconsistent capitalization; use 'bio-inspired' throughout.
  3. [Page header] The page header displays an arXiv identifier (2508.16213v1) that does not match the paper's stated identifier (2508.16216). Please verify and correct.
  4. [Abstract / References] The source of the device data (iontronic organic memtransistor measurements) and the spline fitting procedure are not referenced or described. If this is presented in an appendix, it is not readable in the submitted file.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the device-derived kernels are an external input, and no fitted parameter is renamed as a prediction.

full rationale

The paper's central claim—SADP with spline kernels from memtransistor measurements outperforms STDP on MNIST/Fashion-MNIST—rests on transferring a kernel derived from independent device experiments. Nothing in the provided text shows the kernel parameters or SADP hyperparameters being fit to the test labels, nor any equation in which the output quantity is definitionally equal to an input. The derivation defines SADP via pre-post spike agreement and compares it against STDP as an external baseline. The only concern (whether the device data themselves were tuned to MNIST) is a data-hygiene question that cannot be established from the paper, and per the rules, suspicion without a quoted reduction is not circularity. There are no load-bearing self-citations or imported uniqueness theorems. Hence score 0.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The abstract introduces SADP as a new learning rule, but it is a method rather than a new entity (particle, force, dimension, etc.). The spline-based kernels are a potential source of fitted free parameters.

free parameters (1)
  • Spline-based kernel parameters = not reported in abstract
    The abstract states the kernels are 'derived from our experimental ... device data', which likely involves fitting to device measurements; their values affect the SADP update.
assumptions (3)
  • domain assumption Cohen's kappa is a valid measure of agreement between pre- and post-synaptic spike trains for driving plasticity.
    The abstract proposes using population-level correlation metrics such as Cohen's kappa without proof that this captures the biologically relevant learning signal.
  • ad hoc to paper The spline kernels from device data transfer to MNIST/Fashion-MNIST classification.
    The abstract assumes kernels fit to device measurements generalize to the learning task; this is untested in the abstract.
  • domain assumption Linear-time complexity and bitwise implementability follow from the rule structure.
    The abstract asserts these properties without derivation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Spike Agreement Dependent Plasticity: A scalable Bio-Inspired learning paradigm for Spiking Neural Networks." pith.science (2026). https://pith.science/paper/JLB7LWZ2

@misc{pith2026250816216,
  author       = {Pith},
  title        = {Pith review of: Spike Agreement Dependent Plasticity: A scalable Bio-Inspired learning paradigm for Spiking Neural Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JLB7LWZ2}},
  note         = {Machine review of arXiv:2508.16216}
}
read the original abstract

We introduce Spike Agreement Dependent Plasticity (SADP), a biologically inspired synaptic learning rule for Spiking Neural Networks (SNNs) that relies on the agreement between pre- and post-synaptic spike trains rather than precise spike-pair timing. SADP generalizes classical Spike-Timing-Dependent Plasticity (STDP) by replacing pairwise temporal updates with population-level correlation metrics such as Cohen's kappa. The SADP update rule admits linear-time complexity and supports efficient hardware implementation via bitwise logic. Empirical results on MNIST and Fashion-MNIST show that SADP, especially when equipped with spline-based kernels derived from our experimental iontronic organic memtransistor device data, outperforms classical STDP in both accuracy and runtime. Our framework bridges the gap between biological plausibility and computational scalability, offering a viable learning mechanism for neuromorphic systems.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 21 canonical work pages

  1. [1]

    Bi, M.-m

    G.-q. Bi, M.-m. Poo, Synaptic modifications in cultured hippocampal neurons: dependence on spike timing, synaptic strength, and postsynaptic cell type, Journal of neuroscience 18 (24) (1998) 10464--10472. https://doi.org/10.1523/JNEUROSCI.18-24-10464.1998 doi:10.1523/JNEUROSCI.18-24-10464.1998

    arXiv 1998

  2. [2]

    Markram, J

    H. Markram, J. L \"u bke, M. Frotscher, B. Sakmann, Regulation of synaptic efficacy by coincidence of postsynaptic aps and epsps, Science 275 (5297) (1997) 213--215. https://doi.org/10.1126/science.275.5297.213 doi:10.1126/science.275.5297.213

    doi:10.1126/science.275.5297.213 1997

  3. [3]

    Bi, M.-M

    G.-Q. Bi, M.-M. Poo, Synaptic modification by correlated activity: Hebb's postulate revisited, Annual review of neuroscience 24 (1) (2001) 139--166. https://doi.org/10.1146/annurev.neuro.24.1.139 doi:10.1146/annurev.neuro.24.1.139

    doi:10.1146/annurev.neuro.24.1.139 2001

  4. [4]

    P. J. Sj \"o str \"o m, G. G. Turrigiano, S. B. Nelson, Rate, timing, and cooperativity jointly determine cortical synaptic plasticity, Neuron 32 (6) (2001) 1149--1164. https://doi.org/10.1016/s0896-6273(01)00542-6 doi:10.1016/s0896-6273(01)00542-6

    work page doi:10.1016/s0896-6273(01)00542-6 2001

  5. [5]

    Masquelier, S

    T. Masquelier, S. J. Thorpe, Unsupervised learning of visual features through spike timing dependent plasticity, PLoS Computational Biology 3 (2) (2007) e31. https://doi.org/10.1371/journal.pcbi.0030031 doi:10.1371/journal.pcbi.0030031

    doi:10.1371/journal.pcbi.0030031 2007

  6. [6]

    Ferr \'e , F

    P. Ferr \'e , F. Mamalet, S. J. Thorpe, Unsupervised feature learning with winner-takes-all based stdp, Frontiers in Computational Neuroscience 12 (2018) 24. https://doi.org/10.3389/fncom.2018.00024 doi:10.3389/fncom.2018.00024

    arXiv 2018

  7. [7]

    S. R. Kheradpisheh, M. Ganjtabesh, T. Masquelier, Bio-inspired unsupervised learning of visual features leads to robust invariant object recognition, Neurocomputing 205 (2016) 382--392. https://doi.org/10.1016/j.neucom.2016.04.029 doi:10.1016/j.neucom.2016.04.029

    work page doi:10.1016/j.neucom.2016.04.029 2016

  8. [8]

    A change of direction in pairwise neutrino conversion physics: The effect of collisions

    H. Markram, W. Gerstner, P. J. Sj \"o str \"o m, A history of spike-timing-dependent plasticity, Frontiers in Synaptic Neuroscience 3 (2011) 4. https://doi.org/10.3389/fnsyn.2011.00004 doi:10.3389/fnsyn.2011.00004

    work page Pith review arXiv 2011

Show all 38 references
  1. [9]

    Markram, W

    H. Markram, W. Gerstner, P. J. Sj \"o str \"o m, Spike-timing-dependent plasticity: A comprehensive overview, Frontiers in Synaptic Neuroscience 4 (2012) 2. https://doi.org/10.3389/fnsyn.2012.00002 doi:10.3389/fnsyn.2012.00002

    2012 arXiv

  2. [10]

    Y. Tian, A. Kembay, N. D. Truong, J. K. Eshraghian, O. Kavehei, Beyond pairwise plasticity: Group-level spike synchrony facilitates efficient learning in spiking neural networks, arXivPublished Apr 14 2025 (2025)

    2025

  3. [11]

    L. A. Becker, F. Baccelli, T. Taillefumier, Subthreshold variability of neuronal populations driven by synchronous synaptic inputs, bioRxivPreprint (March 2025). https://doi.org/10.1101/2025.03.16.643547 doi:10.1101/2025.03.16.643547

    2025 doi

  4. [12]

    Foncelle, A

    A. Foncelle, A. Mendes, et al., Modulation of spike-timing dependent plasticity: Towards the inclusion of a third factor in computational models, Frontiers in Computational Neuroscience 12 (2018) 49. https://doi.org/10.3389/fncom.2018.00049 doi:10.3389/fncom.2018.00049

    2018

  5. [13]

    Clopath, L

    C. Clopath, L. B \"u sing, E. Vasilaki, W. Gerstner, Connectivity reflects coding: a model of voltage-based stdp with homeostasis, Nature neuroscience 13 (3) (2010) 344--352. https://doi.org/10.1038/nn.2479 doi:10.1038/nn.2479

    2010 doi

  6. [14]

    Pfister, W

    J.-P. Pfister, W. Gerstner, Triplets of spikes in a model of spike timing-dependent plasticity, Journal of Neuroscience 26 (38) (2006) 9673--9682. https://doi.org/10.1523/JNEUROSCI.1425-06.2006 doi:10.1523/JNEUROSCI.1425-06.2006

    2006 doi

  7. [15]

    D. O. Hebb, The Organization of Behavior, Wiley, 1949

    1949

  8. [16]

    Oja, A simplified neuron model as a principal component analyzer, Journal of mathematical biology 15 (3) (1982) 267--273

    E. Oja, A simplified neuron model as a principal component analyzer, Journal of mathematical biology 15 (3) (1982) 267--273. https://doi.org/10.1007/BF00275687 doi:10.1007/BF00275687

    1982 doi

  9. [17]

    E. L. Bienenstock, L. N. Cooper, P. W. Munro, Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex, Journal of Neuroscience 2 (1) (1982) 32--48. https://doi.org/10.1523/JNEUROSCI.02-01-00032.1982 doi:10.1523/JNEUR...

    1982

  10. [18]

    Gerstner, W

    W. Gerstner, W. M. Kistler, R. Naud, L. Paninski, Neuronal Dynamics: From Single Neurons to Networks and Models of Cognition, Cambridge University Press, 2014. https://doi.org/10.1017/CBO9781107447615 doi:10.1017/CBO9781107447615

    2014 doi

  11. [19]

    S. R. Kheradpisheh, M. Ganjtabesh, S. J. Thorpe, T. Masquelier, Stdp-based spiking deep convolutional neural networks for object recognition, Neural Networks 99 (2018) 56--67. https://doi.org/10.1016/j.neunet.2017.12.005 doi:10.1016/j.neunet.2017.12.005

    2018 doi

  12. [20]

    M. Dong, X. Huang, B. Xu, Unsupervised speech recognition through spike-timing-dependent plasticity in a convolutional spiking neural network, PLoS ONE 13 (11) (2018) e0204596. https://doi.org/10.1371/journal.pone.0204596 doi:10.1371/journal.pone.0204596

    2018 doi

  13. [21]

    S. Song, K. D. Miller, L. F. Abbott, Competitive hebbian learning through spike-timing-dependent synaptic plasticity, Nature Neuroscience 3 (9) (2000) 919--926. https://doi.org/10.1038/78829 doi:10.1038/78829

    2000 doi

  14. [22]

    X. Yang, B. Doiron, Causal spike timing dependent plasticity prevents assembly fusion in recurrent networks https://arxiv.org/abs/2501.09296, aRxiv (2025). ://arxiv.org/abs/2501.09296

    2025 arXiv

  15. [23]

    Chawla, E

    D. Chawla, E. D. Lumer, K. J. Friston, The relationship between synchronization among neuronal populations and their mean activity levels, Neural Computation 11 (6) (1999) 1389--1411. https://doi.org/10.1162/089976699300016287 doi:10.1162/089976699300016287

    1999 doi

  16. [24]

    Nessler, M

    B. Nessler, M. Pfeiffer, L. Buesing, W. Maass, Bayesian computation emerges in generic cortical microcircuits through spike-timing-dependent plasticity, PLoS Computational Biology 9 (4) (2013) e1003037. https://doi.org/10.1371/journal.pcbi.1003037 doi:10.1371/journal.pcbi.1003037

    2013 doi

  17. [25]

    Habenschuss, H

    S. Habenschuss, H. Puhr, W. Maass, Emergence of optimal decoding of population codes through stdp, Neural Computation 25 (6) (2013) 1371--1407. https://doi.org/10.1162/NECO_a_00446 doi:10.1162/NECO_a_00446

    2013 doi

  18. [26]

    Sagar, K

    S. Sagar, K. Udaya Mohanan, S. Cho, et al., Emulation of synaptic functions with low voltage organic memtransistor for hardware oriented neuromorphic computing https://doi.org/10.1038/s41598-022-07505-9, Scientific Reports 12 (1) (2022) 3808. https://doi.org/10.1038/s41598-022...

    2022 doi

  19. [27]

    H. J. Lee, J. H. Hwang, K. B. Choi, S.-G. Jung, K. N. Kim, Y. S. Shim, C. H. Park, Y. W. Park, B.-K. Ju, Effective indium-doped zinc oxide buffer layer on silver nanowires for electrically highly stable, flexible, transparent, and conductive composite electrodes https://doi.or...

    2013 doi

  20. [28]

    Mukherjee, et al., Superionic rubidium silver iodide gated low voltage synaptic transistor https://doi.org/10.1063/5.0069478, Applied Physics Letters 119 (25) (2021) 253502

    A. Mukherjee, et al., Superionic rubidium silver iodide gated low voltage synaptic transistor https://doi.org/10.1063/5.0069478, Applied Physics Letters 119 (25) (2021) 253502. https://doi.org/10.1063/5.0069478 doi:10.1063/5.0069478 . ://doi.org/10.1063/5.0069478

    2021 doi

  21. [29]

    Sagar, A

    S. Sagar, A. Dey, B. C. Das, Unconventional redox-active gate dielectrics to fabricate high performance organic thin-film transistors https://doi.org/10.1021/acsaelm.9b00522, ACS Applied Electronic Materials 1 (11) (2019) 2314--2324. https://doi.org/10.1021/acsaelm.9b00522 doi...

    2019 doi

  22. [30]

    van de Burgt, A

    Y. van de Burgt, A. Melianas, S. T. Keene, et al., Organic electronics for neuromorphic computing https://doi.org/10.1038/s41928-018-0103-3, Nature Electronics 1 (7) (2018) 386--397. https://doi.org/10.1038/s41928-018-0103-3 doi:10.1038/s41928-018-0103-3 . ://doi.org/10.1038/s...

    2018 doi

  23. [31]

    Hazan, D

    H. Hazan, D. J. Saunders, H. Khan, D. Patel, D. T. Sanghavi, H. T. Siegelmann, R. Kozma, Bindsnet: A machine learning-oriented spiking neural networks library in python, Frontiers in Neuroinformatics 12 (2018) 89. https://doi.org/10.3389/fninf.2018.00089 doi:10.3389/fninf.2018.00089

    2018

  24. [32]

    Davies, N

    M. Davies, N. Srinivasa, T.-M. Lin, G. Chinya, Y. Cao, S. Choday, G. Deal, P. Dimoulas, B. Felt, A. Fiebig, et al., Loihi: A neuromorphic manycore processor with on-chip learning, IEEE Micro 38 (1) (2018) 82--99. https://doi.org/10.1109/MM.2018.112130359 doi:10.1109/MM.2018.112130359

    2018

  25. [33]

    P. U. Diehl, M. Cook, Unsupervised learning of digit recognition using spike-timing-dependent plasticity, Frontiers in Computational Neuroscience 9 (2015) 99. https://doi.org/10.3389/fncom.2015.00099 doi:10.3389/fncom.2015.00099

    2015

  26. [34]

    Morrison, M

    A. Morrison, M. Diesmann, W. Gerstner, Phenomenological models of synaptic plasticity based on spike timing, Biological Cybernetics 98 (6) (2008) 459--478. https://doi.org/10.1007/s00422-008-0233-1 doi:10.1007/s00422-008-0233-1

    2008 doi

  27. [35]

    Stimberg, R

    M. Stimberg, R. Brette, D. F. Goodman, Brian 2, an intuitive and efficient neural simulator, eLife 8 (2019) e47314. https://doi.org/10.7554/eLife.47314 doi:10.7554/eLife.47314

    2019 doi

  28. [36]

    Clopath, W

    C. Clopath, W. Gerstner, Voltage and spike timing interact in stdp – a unified model, Frontiers in Synaptic Neuroscience 2 (2010) 25. https://doi.org/10.3389/fnsyn.2010.00025 doi:10.3389/fnsyn.2010.00025

    2010 arXiv

  29. [37]

    Linsker, From basic network principles to neural architecture: Emergence of orientation columns, Proceedings of the National Academy of Sciences 83 (21) (1986) 8779--8783

    R. Linsker, From basic network principles to neural architecture: Emergence of orientation columns, Proceedings of the National Academy of Sciences 83 (21) (1986) 8779--8783. https://doi.org/10.1073/pnas.83.22.8779 doi:10.1073/pnas.83.22.8779

    1986 doi

  30. [38]

    K. D. Miller, D. J. MacKay, The role of constraints in hebbian learning, Neural Computation 6 (1) (1994) 100--126. https://doi.org/10.1162/neco.1994.6.1.100 doi:10.1162/neco.1994.6.1.100

    1994 doi

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.