Pith. sign in

REVIEW 3 major objections 7 minor 67 references

Vertically Integrated Dual-memtransistor Enabled Reconfigurable Heterosynaptic Sensorimotor Networks and In-memory Neuromorphic Computing

T0 review · 3 major / 7 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read A single vertically stacked MoS2/In2Se3 transistor can act as two coupled synapses and a Boolean logic gate, the paper argues.

desk verdict The device work is real and the vertical dual-memtransistor architecture is new, but the headline learning accuracies come from idealized crossbar simulations, not from a fabricated array. read the letter →

arxiv 2412.10757 v1 pith:E444M4JV submitted 2024-12-14 physics.app-ph

classification physics.app-ph
keywords 2DmaterialsvanderWaalsheterostructureferroelectricsemiconductorMoS2/In2Se3memtransistorheterosynapticplasticityin-memoryneuromorphiccomputingreconfigurableBooleanlogic
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper claims that a single vertically stacked transistor, the VSFET, built from a MoS2 channel on top of a ferroelectric-semiconductor In2Se3 channel on hBN, can serve as two coupled synapse devices accessed by one gate. The authors report that gate-controlled ferroelectric polarization of In2Se3 gives the MoS2 channel memtransistor behavior suitable for homosynaptic plasticity, with measured potentiation and depression nonlinearities of $\alpha_p = 1.74$ and $\alpha_d = -0.07$, and that this behavior yields 90.06% test accuracy in a simulated MNIST classifier after five epochs and 96.2% accuracy in a spiking network on Fisher's iris after two epochs. They further report that simultaneously measuring both channels realizes heterosynaptic cooperation and competition, which they use to mimic the gill-withdrawal reflex sensitization and habituation of Aplysia, and that the same device can be reconfigured as NOT and NOR Boolean gates. If correct, a single area-efficient device could serve as memory, two coupled synapses, and logic, a building block for in-memory neuromorphic computing.

What carries the argument

The load-bearing object is the vertically stratified field-effect transistor (VSFET): a bottom Au gate, an hBN dielectric, a thick n-type In2Se3 ferroelectric semiconductor channel, and a thin n-type MoS2 channel stacked on top, with source/drain contacts on both layers so each layer can be read as an independent transistor while sharing one gate. The out-of-plane ferroelectric polarization of In2Se3 is what stores memory: sweeping the gate beyond the coercive field flips the polarization, and the bound charge at the In2Se3/MoS2 interface modulates the MoS2 channel conductance, producing clockwise hysteresis whose memory window grows with sweep range. The in-plane polarization of the same In2Se3 layer modulates the MoS2 output conductance through Schottky-barrier changes at the contacts, producing pinched resistive hysteresis loops at zero gate bias. Because both channels share the gate, one gate pulse updates two coupled conductances at once; that shared-gate coupling is the mechanism behind the heterosynaptic cooperation and competition, and the same stack is reconfigured as a NOT or NOR gate by choosing which terminals serve as inputs and reading the net source current as the output.

What would settle it

Fabricate a small crossbar array of MoS2/In2Se3 VSFETs, measure the conductance update of every cell across repeated potentiation/depression cycles, and run the same MNIST and iris simulations using the measured array-level nonlinearity and variation; if the 90.06% and 96.2% accuracies collapse or fail to reproduce, the single-device claims stand but the array-level neuromorphic claim falls.

Watch

Extended reading notes

Core claim

The central claim is that stacking a non-ferroelectric 2D semiconductor (MoS2) on a ferroelectric 2D semiconductor (In2Se3) creates two electrostatically coupled memtransistors within one vertical footprint, and that this coupling can be harnessed for both synaptic and logic functions. The out-of-plane ferroelectric polarization of In2Se3 modulates the MoS2 channel conductance, producing the gate-controlled hysteresis that underlies homosynaptic plasticity, while the in-plane polarization of the same In2Se3 layer produces pinched resistive output hysteresis at zero gate bias. Because both channel layers share a single gate, the two synapses can be measured simultaneously and show cooperative or competitive plasticity depending on the drain biases, and the same device can be reconfigured as a NOT or NOR gate by choosing which terminals are inputs.

Load-bearing premise

The simulated neural-network accuracies assume that a crossbar array of VSFETs will behave exactly like the single measured device, with no device-to-device variation and ideal interconnects, and no array was fabricated to test this.

Editorial extensions

If this is right

  • A single VSFET cell could replace two separate synapse devices in a crossbar array, roughly halving the area per synapse while keeping both conductances coupled.
  • The measured potentiation and depression nonlinearity, with $\alpha_p - \alpha_d = 1.81$, is low enough that a five-epoch, one-bit-update multilayer perceptron reaches 90.06% on MNIST, so the device offers a route to simple peripheral circuits.
  • The simultaneously read two channels support cooperative and competitive plasticity in one device, which could help balance and refine synaptic weights during learning.
  • The same device can switch between synaptic memory and NOT/NOR logic, enabling a logic-in-memory architecture without extra area.

Reading between the lines

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

  • Beyond the paper, if the single-device behavior survives in an array, the vertical stacking could push synapse density beyond planar designs because two memtransistors occupy one footprint; a fabricated crossbar would be the natural test.
  • The shared-gate coupling means one input pulse updates two weights simultaneously, which could enable training schemes where correlated synapses are updated together rather than independently.
  • The zero-gate-bias resistive switching from in-plane polarization suggests the VSFET could also be read as a two-terminal memory cell when the gate is idle, expanding its role beyond synaptic emulation.
  • Because the measured learning accuracies are simulation-based, a near-term testable extension is to build a small VSFET crossbar and measure classification accuracy under real device-to-device variation.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 7 minor

Summary. The paper reports a vertically stacked MoS2/In2Se3 field-effect transistor (VSFET) in which a top MoS2 channel and a bottom ferroelectric-semiconductor In2Se3 channel share a common gate and are electrostatically coupled. The authors demonstrate hysteretic transfer and output characteristics, attribute the hysteresis to coupled out-of-plane and in-plane ferroelectric polarization of In2Se3, and use the gate-controlled MoS2 channel to emulate homosynaptic plasticity (potentiation/depression and STDP). From measured P/D and STDP curves they simulate supervised MNIST and unsupervised iris classification accuracies of 90.06% and 96.2%, respectively. Simultaneous measurements of both channels are used to demonstrate heterosynaptic cooperation and competition, which are mapped onto a conceptual Aplysia gill-withdrawal model. Finally, the device is reconfigured to implement NOT and NOR logic gates.

Significance. If the claims are properly scoped, the VSFET is a useful building block for in-memory neuromorphic computing: it is area-efficient, combines two coupled synapses in a vertical stack, shows both homosynaptic and heterosynaptic plasticity, and can be reconfigured as logic gates. The strength of the work lies in the direct device-level measurements, the inclusion of C-MoS2 and C-In2Se3 control devices that support the ferroelectric-coupling attribution, the demonstration of simultaneous access to two coupled channels, and the use of measured P/D and STDP data as inputs to network simulations rather than purely abstract model parameters. The main weakness is that the headline 'on-chip learning' accuracies are not hardware results but single-device-to-simulation extrapolations, so the significance depends on whether array-level variability is small or the claims are appropriately qualified.

major comments (3)
  1. [Homosynaptic Plasticity and Neuromorphic In-Memory Computing, Figure 4] The abstract and main text claim 'high accuracy supervised and unsupervised on-chip learning' (90.06% MNIST, 96.2% iris), but these numbers are not obtained from a fabricated crossbar array; they are simulations that use the measured P/D and STDP curves from a single device. The manuscript itself states that 'the P/D parameters extracted from Figure 4b were used to model the hardware implementation of a NN circuit utilizing MoS2/In2Se3 vdW VSFET arrays in crossbar architecture,' and no array-level measurements or multi-device statistics are reported. The learning-accuracy claims therefore rest on the untested assumption that the single measured device is representative of an entire array with no device-to-device variation, interconnect resistance, or write noise. To support the central claim, the authors should either present array or statistical-ensemble measurements, or explicitly re-label these results as simulations and qualify the 'on-chip' language in the abstract.
  2. [Homosynaptic Plasticity and Neuromorphic In-Memory Computing, Table 2 and Figure 4d] The relationship between the reported test accuracies is inconsistent and unexplained. Figure 4d reports 90.06% for 'on-chip learning' using the incremental-amplitude pulse scheme, while Table 2 lists 97.21% for the same scheme under 'on-chip inference.' If the Table 2 numbers refer to inference after offline training with ideal weights mapped to the device, that should be stated; if they refer to on-chip training, the discrepancy with Figure 4d is substantial and needs a reason. Without clarification, the central accuracy numbers cannot be interpreted.
  3. [Homosynaptic Plasticity and Neuromorphic In-Memory Computing, Figure 4f] The 96.2% unsupervised SNN accuracy on Fisher's iris is presented without the network architecture, neuron model, input encoding, or spike-timing simulation details; the reader is only referred to prior work [44]. Given that the measured STDP time constants are approximately 113 ms and 337 ms, the SNN simulation must also specify the mapping between physical time and simulation time. Without these details, the unsupervised learning result is not reproducible.
minor comments (7)
  1. [Abstract] The abstract describes the device as showing 'low nonlinearity' but the reported potentiation nonlinearity factor is alpha_p = 1.74; please state the numerical values or clarify the comparison baseline.
  2. [Figure 4d caption] The caption refers to 'on-chip learning accuracy numbers' but these numbers come from simulations. Please use 'simulated' or 'modeled' in the caption and throughout the text.
  3. [Supporting Information 7] The mapping from the fitted parameters A_P and A_D to the nonlinearity factors alpha_p and alpha_d is not shown explicitly; please provide the full formula or a reference to the exact NeuroSim documentation page.
  4. [Energy per spike, Homosynaptic Plasticity section] The equation E_perspike = I_PSC * V_D * t_spike uses the drain-side voltage, but the presynaptic spike is applied at the gate; please justify why gate leakage current is negligible and clarify why the drain-side energy is the relevant metric.
  5. [Biomimetic Heterosynaptic Interactions in Sea Mollusk] The text says 'the HT-MoS2 and HB-MoS2 synapses' but the bottom channel is In2Se3, not MoS2; this should read 'HT-MoS2 and HB-In2Se3.'
  6. [Table 1, Conclusion] In the comparison table, 'Fisher's Irish' should be 'Fisher's Iris.'
  7. [References [45-47]] References [45-47] are URLs of Wikimedia Commons images; these are not standard citations for a scientific journal and should be replaced with proper data or literature citations.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the learning accuracies are network simulations fed by measured P/D and STDP curves, not predictions forced by the model's own definitions.

full rationale

The paper's derivation chain is a sequence of measurements followed by simulations and biological analogies. Measured transfer/output hysteresis and PFM establish ferroelectric coupling; measured P/D responses (Fig. 4b) and STDP curves (Fig. 4e) are used as weight-update characteristics in standard backpropagation and spiking-neural-network frameworks (refs 39, 40) to obtain the reported MNIST and Fisher Iris accuracies. The accuracy numbers are conditional simulations of an ideal crossbar, not a hidden return of the fitting inputs: the same measured curves could feed many network models and yield different accuracies, and the network output is not used to define the device parameters. The references to Kaushik et al. and Yadav et al. are self-citations by co-author Bhowmik, but they are invoked only for simulator/training-algorithm details and are not the source of the device's measured nonlinearity or STDP constants. The heterosynaptic cooperation/competition and Aplysia mimicry are interpretive labels applied to simultaneous measurements of two coupled channels sharing a gate; they involve no quasi-theoretical prediction that reduces to its own inputs. The main weakness is external validity: no crossbar array was fabricated, and the network accuracy extrapolates from one device, an extrapolation/validation gap rather than circular reasoning. Overall score 0.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claims rest on four fitted parameters (potentiation and depression nonlinearity factors, two STDP time constants) and on domain assumptions that include extrapolating a single device to an array and equating two stacked channels with two biological synapses. No new physical entities are introduced.

free parameters (4)
  • Potentiation nonlinearity factor alpha_p = 1.74
    Fitted to the measured potentiation curve via the Neurosim mapping (SI Section 7); used in the ANN simulation.
  • Depression nonlinearity factor alpha_d = -0.07
    Fitted to the measured depression curve via the Neurosim mapping; used in the ANN simulation.
  • STDP potentiation time constant tau+ = 113 ms
    Exponential fit to measured STDP data for positive Delta t, reported in the STDP section.
  • STDP depression time constant tau- = 337 ms
    Exponential fit to measured STDP data for negative Delta t.
assumptions (5)
  • domain assumption Hysteresis in HT-MoS2 originates from ferroelectric polarization of In2Se3 rather than charge trapping or interface defects.
    Used in the Device Physics section to explain MW expansion with VGmax; supported by the C-MoS2 control and PFM, but trapping is not explicitly excluded.
  • domain assumption A crossbar array of VSFETs will reproduce the single-device P/D and transfer behavior with negligible device-to-device variation.
    Invoked when P/D parameters from one device are used to model MNIST accuracy in Figure 4c-d; no array was fabricated.
  • domain assumption Reading the two stacked channels simultaneously represents heterosynaptic plasticity between two synapses.
    The biological mapping in Figures 5-6 treats the two channels as two synapses; the shared gate is a simplification relative to biology.
  • domain assumption alpha-In2Se3 retains its ferroelectric and correlated IP-OOP polarization when covered by MoS2.
    PFM is performed on In2Se3 flakes (Figure S1), and device behavior is consistent, but no direct PFM on the final stacked device is shown.
  • standard math Standard backpropagation and STDP learning rules apply to the simulated networks.
    The ANN and SNN simulations rely on textbook learning algorithms; this is not a source of circularity.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Vertically Integrated Dual-memtransistor Enabled Reconfigurable Heterosynaptic Sensorimotor Networks and In-memory Neuromorphic Computing." pith.science (2026). https://pith.science/paper/E444M4JV

@misc{pith2026241210757,
  author       = {Pith},
  title        = {Pith review of: Vertically Integrated Dual-memtransistor Enabled Reconfigurable Heterosynaptic Sensorimotor Networks and In-memory Neuromorphic Computing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/E444M4JV}},
  note         = {Machine review of arXiv:2412.10757}
}
abstract

Neuromorphic in-memory computing requires area-efficient architecture for seamless and low latency parallel processing of large volumes of data. Here, we report a compact, vertically integrated/stratified field-effect transistor (VSFET) consisting of a 2D non-ferroelectric MoS$_2$ FET channel stacked on a 2D ferroelectric In$_2$Se$_3$ FET channel. Electrostatic coupling between the ferroelectric and non-ferroelectric semiconducting channels results in hysteretic transfer and output characteristics of both FETs. The gate-controlled MoS$_2$ memtransistor is shown to emulate homosynaptic plasticity behavior with low nonlinearity, low epoch, and high accuracy supervised (ANN - artificial neural network) and unsupervised (SNN - spiking neural network) on-chip learning. Further, simultaneous measurements of the MoS$_2$ and In$_2$Se$_3$ transistor synapses help realize complex heterosynaptic cooperation and competition behaviors. These are shown to mimic advanced sensorimotor neural network-controlled gill withdrawal reflex sensitization and habituation of a sea mollusk (Aplysia) with ultra-low power consumption. Finally, we show logic reconfigurability of the VSFET to realize Boolean gates thereby adding significant design flexibility for advanced computing technologies.

Figures

Figures reproduced from arXiv: 2412.10757 by the authors.

Figure 1
Figure 1. Device structure and Raman spectra. (a) 3D schematic of MoS [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. DC electrical characteristics of VSFET. (a) Comparison of transfer characteristics [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Device physics for |VGmax| dependent ID - VG hysteresis in HT-MoS2 FET. (a) The transfer curve of HT-MoS2 FET on the left has four marked points, 1, 2, 3, and 4, representing four operating points of the device in a clockwise direction for |VGmax| = 10 V. Energy band diagrams for (b) point 2 and (c) point 4 illustrate the change in electrostatics, spatial electric field, and charge distribution corresponding to a ch… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Neuromorphic in-memory computing using HT-MoS [PITH_FULL_IMAGE:figures/full_fig_p015_4.png]
Figure 5
Figure 5. Figure 5: Heterosynaptic interactions utilizing VSFET. (a) Schematics showing two inter [PITH_FULL_IMAGE:figures/full_fig_p017_5.png]
Figure 6
Figure 6. Figure 6: Biomimetic heterosynaptic response and defensive learning behavior of sea mol [PITH_FULL_IMAGE:figures/full_fig_p019_6.png]
Figure 7
Figure 7. Figure 7: Boolean logic gates using VSFET. (a) 2D device cross-section biasing schematic [PITH_FULL_IMAGE:figures/full_fig_p020_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

67 extracted references · 65 canonical work pages

  1. [44]

    Spike time dependent plasticity (STDP) enabled learning in spiking neural networks using domain wall based synapses and neurons

    Sahu, U.; Pandey, A.; Goyal, K.; Bhowmik, D. Spike time dependent plasticity (STDP) enabled learning in spiking neural networks using domain wall based synapses and neurons. AIP Advances 2019, 9

  2. [1]

    Ultra-low power neuromorphic obstacle detection using a two-dimensional materials-based subthreshold transistor

    Thakar, K.; Rajendran, B.; Lodha, S. Ultra-low power neuromorphic obstacle detection using a two-dimensional materials-based subthreshold transistor. npj 2D Materials and Applications 2023, 7, 68

  3. [2]

    L.; Bhattacharya, S.; Sreejith, S.; Kollem, S.; Rebelli, S.; Tayal, S.; Mounika, B

    Ajayan, J.; Mohankumar, P.; Nirmal, D.; Joseph, L. L.; Bhattacharya, S.; Sreejith, S.; Kollem, S.; Rebelli, S.; Tayal, S.; Mounika, B. Ferroelectric field effect transistors (FeFETs): advancements, challenges and exciting prospects for next generation non-volatile memory (NVM) applications. Materials Today Communications 2023, 35, 105591

  4. [3]

    Optoelectrical Molybdenum Disulfide (MoS2)—Ferroelectric Memories

    Lipatov, A.; Sharma, P.; Gruverman, A.; Sinitskii, A. Optoelectrical Molybdenum Disulfide (MoS2)—Ferroelectric Memories. ACS nano 2015, 9, 8089--8098

  5. [4]

    Advanced Electronic Materials 2019, 5, 1800600

    Tian, B.; Liu, L.; Yan, M.; Wang, J.; Zhao, Q.; Zhong, N.; Xiang, P.; Sun, L.; Peng, H.; Shen, H.; others A robust artificial synapse based on organic ferroelectric polymer. Advanced Electronic Materials 2019, 5, 1800600

  6. [5]

    Chen, L.; Wang, L.; Peng, Y.; Feng, X.; Sarkar, S.; Li, S.; Li, B.; Liu, L.; Han, K.; Gong, X.; others A van der Waals synaptic transistor based on ferroelectric Hf0. 5Zr0. 5O2 and 2D tungsten disulfide. Advanced Electronic Materials 2020, 6, 2000057

  7. [6]

    R.; Soliman, M.; Taniguchi, T.; Watanabe, K.; Doudin, B.; Ouerghi, A.; others Reconfigurable multifunctional van der Waals ferroelectric devices and logic circuits

    Ram, A.; Maity, K.; Marchand, C.; Mahmoudi, A.; Kshirsagar, A. R.; Soliman, M.; Taniguchi, T.; Watanabe, K.; Doudin, B.; Ouerghi, A.; others Reconfigurable multifunctional van der Waals ferroelectric devices and logic circuits. ACS nano 2023, 17, 21865--21877

  8. [7]

    Wang, W.; Meng, Y.; Wang, W.; Zhang, Y.; Li, B.; Yan, Y.; Gao, B.; Ho, J. C. 2D ferroelectric materials: Emerging paradigms for next-generation ferroelectronics. Materials Today Electronics 2023, 6, 100080

Show all 67 references
  1. [8]

    H.; Ju, J

    Baek, S.; Yoo, H. H.; Ju, J. H.; Sriboriboon, P.; Singh, P.; Niu, J.; Park, J.-H.; Shin, C.; Kim, Y.; Lee, S. Ferroelectric field-effect-transistor integrated with ferroelectrics heterostructure. Advanced Science 2022, 9, 2200566

  2. [9]

    L.; Li, X.; Yu, P.; Lin, J.; Wang, X.; Zhou, J.; Wang, H.; He, H.; others Room-temperature ferroelectricity in CuInP2S6 ultrathin flakes

    Liu, F.; You, L.; Seyler, K. L.; Li, X.; Yu, P.; Lin, J.; Wang, X.; Zhou, J.; Wang, H.; He, H.; others Room-temperature ferroelectricity in CuInP2S6 ultrathin flakes. Nature communications 2016, 7, 1--6

  3. [10]

    Towards two-dimensional van der Waals ferroelectrics

    Wang, C.; You, L.; Cobden, D.; Wang, J. Towards two-dimensional van der Waals ferroelectrics. Nature Materials 2023, 22, 542--552

  4. [11]

    V.; Lodha, S

    Ghosh, S.; Varghese, A.; Jawa, H.; Yin, Y.; Medhekar, N. V.; Lodha, S. Polarity-tunable photocurrent through band alignment engineering in a high-speed WSe2/SnSe2 diode with large negative responsivity. ACS nano 2022, 16, 4578--4587

  5. [12]

    V.; Ghosh, S.; Lodha, S

    Varghese, A.; Saha, D.; Thakar, K.; Jindal, V.; Ghosh, S.; Medhekar, N. V.; Ghosh, S.; Lodha, S. Near-direct bandgap WSe2/ReS2 type-II pn heterojunction for enhanced ultrafast photodetection and high-performance photovoltaics. Nano letters 2020, 20, 1707--1717

  6. [13]

    Belianinov, A.; He, Q.; Dziaugys, A.; Maksymovych, P.; Eliseev, E.; Borisevich, A.; Morozovska, A.; Banys, J.; Vysochanskii, Y.; Kalinin, S. V. CuInP2S6 room temperature layered ferroelectric. Nano letters 2015, 15, 3808--3814

  7. [14]

    K.; Gao, S.; Qiu, G.; Qin, J.; Duan, Y.; Jian, J.; Niu, C.; Wang, H.; Wu, W.; others A ferroelectric semiconductor field-effect transistor

    Si, M.; Saha, A. K.; Gao, S.; Qiu, G.; Qin, J.; Duan, Y.; Jian, J.; Niu, C.; Wang, H.; Wu, W.; others A ferroelectric semiconductor field-effect transistor. Nature Electronics 2019, 2, 580--586

  8. [15]

    ACS nano 2023, 17, 6095--6102

    Liao, J.; Wen, W.; Wu, J.; Zhou, Y.; Hussain, S.; Hu, H.; Li, J.; Liaqat, A.; Zhu, H.; Jiao, L.; others Van der Waals ferroelectric semiconductor field effect transistor for in-memory computing. ACS nano 2023, 17, 6095--6102

  9. [16]

    J.; Ji, J.-R.; Zhang, J.-L.; Sessi, P.; Barraza-Lopez, S.; Parkin, S

    Chang, K.; Küster, F.; Miller, B. J.; Ji, J.-R.; Zhang, J.-L.; Sessi, P.; Barraza-Lopez, S.; Parkin, S. S. Microscopic manipulation of ferroelectric domains in SnSe monolayers at room temperature. Nano letters 2020, 20, 6590--6597

  10. [17]

    Purely in-plane ferroelectricity in monolayer SnS at room temperature

    Higashitarumizu, N.; Kawamoto, H.; Lee, C.-J.; Lin, B.-H.; Chu, F.-H.; Yonemori, I.; Nishimura, T.; Wakabayashi, K.; Chang, W.-H.; Nagashio, K. Purely in-plane ferroelectricity in monolayer SnS at room temperature. Nature communications 2020, 11, 2428

  11. [18]

    Nano letters 2019, 19, 5109--5117

    Bao, Y.; Song, P.; Liu, Y.; Chen, Z.; Zhu, M.; Abdelwahab, I.; Su, J.; Fu, W.; Chi, X.; Yu, W.; others Gate-tunable in-plane ferroelectricity in few-layer SnS. Nano letters 2019, 19, 5109--5117

  12. [19]

    A.; Martin, L

    Xiao, J.; Zhu, H.; Wang, Y.; Feng, W.; Hu, Y.; Dasgupta, A.; Han, Y.; Wang, Y.; Muller, D. A.; Martin, L. W.; others Intrinsic two-dimensional ferroelectricity with dipole locking. Physical review letters 2018, 120, 227601

  13. [20]

    Nano letters 2018, 18, 1253--1258

    Cui, C.; Hu, W.-J.; Yan, X.; Addiego, C.; Gao, W.; Wang, Y.; Wang, Z.; Li, L.; Cheng, Y.; Li, P.; others Intercorrelated in-plane and out-of-plane ferroelectricity in ultrathin two-dimensional layered semiconductor In2Se3. Nano letters 2018, 18, 1253--1258

  14. [21]

    Laterally gated ferroelectric field effect transistor (LG-FeFET) using -In2Se3 for stacked in-memory computing array

    Park, S.; Lee, D.; Kang, J.; Choi, H.; Park, J.-H. Laterally gated ferroelectric field effect transistor (LG-FeFET) using -In2Se3 for stacked in-memory computing array. Nature Communications 2023, 14, 6778

  15. [22]

    W.; Zhou, P

    Wang, S.; Liu, L.; Gan, L.; Chen, H.; Hou, X.; Ding, Y.; Ma, S.; Zhang, D. W.; Zhou, P. Two-dimensional ferroelectric channel transistors integrating ultra-fast memory and neural computing. Nature Communications 2021, 12, 53

  16. [23]

    H.; Kim, H

    Kang, S.-J.; Jung, W.; Gwon, O. H.; Kim, H. S.; Byun, H. R.; Kim, J. Y.; Jang, S. G.; Shin, B.; Kwon, O.; Cho, B.; others Photo-Assisted Ferroelectric Domain Control for -In2Se3 Artificial Synapses Inspired by Spontaneous Internal Electric Fields. Small 2024, 20, 2307346

  17. [24]

    M.; Lodha, S

    Kaushik, N.; Nipane, A.; Basheer, F.; Dubey, S.; Grover, S.; Deshmukh, M. M.; Lodha, S. Schottky barrier heights for Au and Pd contacts to MoS2. Applied Physics Letters 2014, 105

  18. [25]

    V.; Lodha, S

    Jawa, H.; Varghese, A.; Ghosh, S.; Sahoo, S.; Yin, Y.; Medhekar, N. V.; Lodha, S. Wavelength-controlled photocurrent polarity switching in BP-MoS2 heterostructure. Advanced Functional Materials 2022, 32, 2112696

  19. [26]

    Interfacial thermal conductance between few to tens of layered-MoS2 and c-Si: Effect of MoS2 thickness

    Yuan, P.; Li, C.; Xu, S.; Liu, J.; Wang, X. Interfacial thermal conductance between few to tens of layered-MoS2 and c-Si: Effect of MoS2 thickness. Acta Materialia 2017, 122, 152--165

  20. [27]

    E.; Heinz, T

    Lee, C.; Yan, H.; Brus, L. E.; Heinz, T. F.; Hone, J.; Ryu, S. Anomalous lattice vibrations of single-and few-layer MoS2. ACS nano 2010, 4, 2695--2700

  21. [28]

    An optoelectronic synapse based on -In2Se3 with controllable temporal dynamics for multimode and multiscale reservoir computing

    Liu, K.; Zhang, T.; Dang, B.; Bao, L.; Xu, L.; Cheng, C.; Yang, Z.; Huang, R.; Yang, Y. An optoelectronic synapse based on -In2Se3 with controllable temporal dynamics for multimode and multiscale reservoir computing. Nature Electronics 2022, 5, 761--773

  22. [29]

    Cross-field optoelectronic modulation via inter-coupled ferroelectricity in 2D In2Se3

    Dutta, D.; Mukherjee, S.; Uzhansky, M.; Koren, E. Cross-field optoelectronic modulation via inter-coupled ferroelectricity in 2D In2Se3. npj 2D Materials and Applications 2021, 5, 81

  23. [30]

    ACS nano 2018, 12, 4976--4983

    Xue, F.; Zhang, J.; Hu, W.; Hsu, W.-T.; Han, A.; Leung, S.-F.; Huang, J.-K.; Wan, Y.; Liu, S.; Zhang, J.; others Multidirection piezoelectricity in mono-and multilayered hexagonal -In2Se3. ACS nano 2018, 12, 4976--4983

  24. [31]

    Wang, L.; Wang, X.; Zhang, Y.; Li, R.; Ma, T.; Leng, K.; Chen, Z.; Abdelwahab, I.; Loh, K. P. Exploring ferroelectric switching in -In2Se3 for neuromorphic computing. Advanced Functional Materials 2020, 30, 2004609

  25. [32]

    H.; Sharma, P.; Yin, Y.; Medhekar, N

    Varghese, A.; Pandey, A. H.; Sharma, P.; Yin, Y.; Medhekar, N. V.; Lodha, S. Electrically Controlled High Sensitivity Strain Modulation in MoS2 Field-Effect Transistors via a Piezoelectric Thin Film on Silicon Substrates. Nano Letters 2024, 24, 8472--8480

  26. [33]

    Xue, F.; He, X.; Wang, Z.; Retamal, J. R. D.; Chai, Z.; Jing, L.; Zhang, C.; Fang, H.; Chai, Y.; Jiang, T.; others Giant ferroelectric resistance switching controlled by a modulatory terminal for low-power neuromorphic in-memory computing. Advanced Materials 2021, 33, 2008709

  27. [34]

    Electrically tunable room temperature hysteresis crossover in underlap MoS2 field-effect transistors

    Jawa, H.; Varghese, A.; Lodha, S. Electrically tunable room temperature hysteresis crossover in underlap MoS2 field-effect transistors. ACS Applied Materials & Interfaces 2021, 13, 9186--9194

  28. [35]

    Mu, Y.; Yang, J.; Xie, G.; Wang, Z.; Guo, B.; Gong, J. R. Homo-type -In2Se3/PdSe2 Ferroelectric van der Waals Heterojunction Photodetectors with High-performance and Broadband. Advanced Functional Materials 2024, 2315543

  29. [36]

    Thickness-dependent band gap of -In2Se3: from electron energy loss spectroscopy to density functional theory calculations

    Lyu, F.; Sun, Y.; Yang, Q.; Tang, B.; Li, M.; Li, Z.; Sun, M.; Gao, P.; Ye, L.-H.; Chen, Q. Thickness-dependent band gap of -In2Se3: from electron energy loss spectroscopy to density functional theory calculations. Nanotechnology 2020, 31, 315711

  30. [37]

    Nature communications 2019, 10, 2639

    Elias, C.; Valvin, P.; Pelini, T.; Summerfield, A.; Mellor, C.; Cheng, T.; Eaves, L.; Foxon, C.; Beton, P.; Novikov, S.; others Direct band-gap crossover in epitaxial monolayer boron nitride. Nature communications 2019, 10, 2639

  31. [38]

    Hexagonal boron nitride is an indirect bandgap semiconductor

    Cassabois, G.; Valvin, P.; Gil, B. Hexagonal boron nitride is an indirect bandgap semiconductor. Nature photonics 2016, 10, 262--266

  32. [39]

    Synapse cell optimization and back-propagation algorithm implementation in a domain wall synapse based crossbar neural network for scalable on-chip learning

    Kaushik, D.; Sharda, J.; Bhowmik, D. Synapse cell optimization and back-propagation algorithm implementation in a domain wall synapse based crossbar neural network for scalable on-chip learning. Nanotechnology 2020, 31, 364004

  33. [40]

    S.; Gupta, P.; Holla, A.; Ali Khan, K

    Yadav, R. S.; Gupta, P.; Holla, A.; Ali Khan, K. I.; Muduli, P. K.; Bhowmik, D. Demonstration of Synaptic Behavior in a Heavy-Metal-Ferromagnetic-Metal-Oxide-Heterostructure-Based Spintronic Device for On-Chip Learning in Crossbar-Array-Based Neural Networks. ACS Applied Elect...

  34. [41]

    S.; Kim, A

    Hwang, Y.; Park, B.; Hwang, S.; Choi, S.-W.; Kim, H. S.; Kim, A. R.; Choi, J. W.; Yoon, J.; Kwon, J.-D.; Kim, Y. A Bioinspired Ultra Flexible Artificial van der Waals 2D-MoS2 Channel/LiSiOx Solid Electrolyte Synapse Arrays via Laser-Lift Off Process for Wearable Adaptive Neuro...

  35. [42]

    Long-term potentiation and long-term depression in experience-dependent plasticity

    Crair, M.; Shah, R. Long-term potentiation and long-term depression in experience-dependent plasticity. 2009,

  36. [43]

    To spike, or when to spike? Current opinion in neurobiology 2014, 25, 134--139

    G \"u tig, R. To spike, or when to spike? Current opinion in neurobiology 2014, 25, 134--139

  37. [45]

    Stang, D. J. Iris setosa 2. https://commons.wikimedia.org/wiki/File:Iris_setosa_2.jpg, 2010; Accessed: 2024-07-01

  38. [46]

    Iris versicolor 3

    Mayfield, F. Iris versicolor 3. https://commons.wikimedia.org/wiki/File:Iris_versicolor_3.jpg, 2007; Accessed: 2024-07-01

  39. [47]

    Richards, F. D. Iris virginica. https://commons.wikimedia.org/wiki/File:Iris_virginica.jpg, 2009; Accessed: 2024-07-01

  40. [48]

    M.; Bazhenov, M.; Volgushev, M

    Chistiakova, M.; Bannon, N. M.; Bazhenov, M.; Volgushev, M. Heterosynaptic plasticity: multiple mechanisms and multiple roles. The Neuroscientist 2014, 20, 483--498

  41. [49]

    Heterosynaptic plasticity in the neocortex

    Chistiakova, M.; Volgushev, M. Heterosynaptic plasticity in the neocortex. Experimental brain research 2009, 199, 377--390

  42. [50]

    R.; Tsimring, K.; Ip, J

    Jenks, K. R.; Tsimring, K.; Ip, J. P. K.; Zepeda, J. C.; Sur, M. Heterosynaptic plasticity and the experience-dependent refinement of developing neuronal circuits. Frontiers in neural circuits 2021, 15, 803401

  43. [51]

    F.; Israely, I

    Ramiro-Cort \'e s, Y.; Hobbiss, A. F.; Israely, I. Synaptic competition in structural plasticity and cognitive function. Philosophical Transactions of the Royal Society B: Biological Sciences 2014, 369, 20130157

  44. [52]

    Zhu, X.; Li, D.; Liang, X.; Lu, W. D. Ionic modulation and ionic coupling effects in MoS2 devices for neuromorphic computing. Nature materials 2019, 18, 141--148

  45. [53]

    Miller, K. D. Synaptic economics: competition and cooperation in synaptic plasticity. Neuron 1996, 17, 371--374

  46. [54]

    Castellucci, V.; Kandel, E. R. Presynaptic facilitation as a mechanism for behavioral sensitization in Aplysia. Science 1976, 194, 1176--1178

  47. [55]

    Kandel, E. R. The molecular biology of memory storage: a dialogue between genes and synapses. Science 2001, 294, 1030--1038

  48. [56]

    Nature Electronics 2020, 3, 383--390

    Pan, C.; Wang, C.-Y.; Liang, S.-J.; Wang, Y.; Cao, T.; Wang, P.; Wang, C.; Wang, S.; Cheng, B.; Gao, A.; others Reconfigurable logic and neuromorphic circuits based on electrically tunable two-dimensional homojunctions. Nature Electronics 2020, 3, 383--390

  49. [57]

    K.; Lee, H.-S.; Bergeron, H.; Balla, I.; Beck, M

    Sangwan, V. K.; Lee, H.-S.; Bergeron, H.; Balla, I.; Beck, M. E.; Chen, K.-S.; Hersam, M. C. Multi-terminal memtransistors from polycrystalline monolayer molybdenum disulfide. Nature 2018, 554, 500--504

  50. [58]

    D.; Huang, R

    Yang, Y.; Yin, M.; Yu, Z.; Wang, Z.; Zhang, T.; Cai, Y.; Lu, W. D.; Huang, R. Multifunctional Nanoionic Devices Enabling Simultaneous Heterosynaptic Plasticity and Efficient In-Memory Boolean Logic. Advanced Electronic Materials 2017, 3, 1700032

  51. [59]

    Versatile functionality of four-terminal TiO2--x memristive devices as artificial synapses for neuromorphic computing

    Miyake, R.; Nagata, Z.; Adachi, K.; Hayashi, Y.; Tohei, T.; Sakai, A. Versatile functionality of four-terminal TiO2--x memristive devices as artificial synapses for neuromorphic computing. ACS Applied Electronic Materials 2022, 4, 2326--2336

  52. [60]

    Heterosynaptic plasticity achieved by highly anisotropic ionic migration in layered LixMoO3 for neuromorphic application

    Qin, J.-K.; Zhu, B.-X.; Wang, C.; Zhu, C.-Y.; Sun, R.-Y.; Zhen, L.; Chai, Y.; Xu, C.-Y. Heterosynaptic plasticity achieved by highly anisotropic ionic migration in layered LixMoO3 for neuromorphic application. Advanced Electronic Materials 2022, 8, 2200721

  53. [61]

    Reconfigurable 2D WSe2-based memtransistor for mimicking homosynaptic and heterosynaptic plasticity

    Ding, G.; Yang, B.; Chen, R.-S.; Mo, W.-A.; Zhou, K.; Liu, Y.; Shang, G.; Zhai, Y.; Han, S.-T.; Zhou, Y. Reconfigurable 2D WSe2-based memtransistor for mimicking homosynaptic and heterosynaptic plasticity. Small 2021, 17, 2103175

  54. [62]

    Brain-inspired structural plasticity through reweighting and rewiring in multi-terminal self-organizing memristive nanowire networks

    Milano, G.; Pedretti, G.; Fretto, M.; Boarino, L.; Benfenati, F.; Ielmini, D.; Valov, I.; Ricciardi, C. Brain-inspired structural plasticity through reweighting and rewiring in multi-terminal self-organizing memristive nanowire networks. Advanced Intelligent Systems 2020, 2, 2000096

  55. [63]

    -lZibkga j(| WAa[ AKC =vȖ^c᭏ ޺o_y7<]|cƌ+zz<j1

    Zheng, C.; Liao, Y.; Xiong, Z.; Zhou, Y.; Han, S.-T. Mimicking the competitive and cooperative behaviors with multi-terminal synaptic memtransistors. Journal of Materials Chemistry C 2020, 8, 6063--6071 mcitethebibliography manuscript.bib000066400000000000000000001114421472723...

  56. [64]

    DNN NeuroSim V2.0 Nonlinearity Norms Documentation

    Contributors, N. DNN NeuroSim V2.0 Nonlinearity Norms Documentation. https://github.com/neurosim/DNN_NeuroSim_V2.0/blob/master/Documents/Nonlinearity-NormA.htm, 2024; Accessed: 26 July 2024

  57. [65]

    Moroz, L. L. Aplysia. Current Biology 2011, 21, R60--R61

  58. [66]

    Habituation and dishabituation of the gill-withdrawal reflex in Aplysia

    Pinsker, H.; Kupfermann, I.; Castellucci, V.; Kandel, E. Habituation and dishabituation of the gill-withdrawal reflex in Aplysia. Science 1970, 167, 1740--1742

  59. [67]

    A.; Li, Q.; Glanzman, D

    Chitwood, R. A.; Li, Q.; Glanzman, D. L. Serotonin facilitates AMPA-type responses in isolated siphon motor neurons of Aplysia in culture. The Journal of physiology 2001, 534, 501 mcitethebibliography supporting.bib0000664000000000000000000010203314727237532012461 0ustar rootr...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.