Pith. sign in

REVIEW 4 major objections 5 minor 14 references

Quantum Elastic Network Models and their Application to Graphene

T0 review · 4 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read A quantum algorithm can encode centimeter-scale graphene vibrations in about 160 logical qubits by mapping elastic network models onto a coupled-oscillator simulation whose cost grows only logarithmically with atom count.

desk verdict A thorough, honest application of Babbush et al.'s oscillator algorithm to graphene ENMs, with a load-bearing flaw in the rippling section: the central-force ENM has zero out-of-plane restoring force. read the letter →

arxiv 2601.05161 v2 pith:ORT25I3U submitted 2026-01-08 quant-ph cond-mat.mtrl-sciphysics.comp-ph

Ioannis Kolotouros , Adithya Sireesh , Stuart Ferguson , Sean Thrasher , Petros Wallden , Julien Michel This is my paper
classification quant-phcond-mat.mtrl-sciphysics.comp-ph
keywords quantumelasticnetworkmodelgraphenesimulationcoupledharmonicoscillatorsblockencodingMaxwell-BoltzmanndiscretizationHamiltonianheattransferout-of-planerippling
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 tries to show that a widely used coarse-grained model of molecular vibrations—the elastic network model, a grid of masses and springs—can be simulated on a quantum computer with resources that grow only logarithmically with the number of atoms. The authors construct quantum elastic network models (QENMs) and instantiate them for a 2D graphene sheet, building every subroutine a prior coupled-oscillator algorithm requires: an efficient connectivity oracle exploiting the lattice's unit-cell structure, a way to load exponentially many initial velocities from a discretized Maxwell-Boltzmann distribution, and a block encoding of the spring Hamiltonian. Their headline estimate is that a 1 cm² graphene sheet—about 3.8 quadrillion carbon atoms—could be represented in roughly 160 logical qubits, whereas classical storage alone would need about 180 petabytes. If this holds, atomistic-scale vibrational simulations of planar materials could move within reach of early fault-tolerant quantum computers. The paper also proposes two concrete experiments: tracking ballistic heat transfer and computing out-of-plane rippling amplitudes.

What carries the argument

The load-bearing construction is the block-encoded Hamiltonian H = [[0, B], [B†, 0]], where B is the weighted incidence matrix of the spring graph and BB† = A is the mass-scaled spring matrix whose evolution solves Newton's equations. The paper's contribution is to make each ingredient efficient for graphene: a connectivity oracle Sa |j,l⟩ → |j,a(j,l)⟩ built from unit-cell shift vectors by quantum adders; a deterministic two-bucket velocity loader that assigns each node a ±σ velocity by a random parity check; and a block encoding of B padded to N × N². Hamiltonian evolution is then implemented by quantum signal processing. The alternative displacement encoding uses the pseudo-inverse B† and

What would settle it

Run a classical harmonic-lattice simulation of a large graphene patch twice, once with initial velocities drawn from the true Maxwell-Boltzmann distribution and once from the two-bucket ±σ distribution at the same kinetic energy, and compare the time-dependent heat-front position and out-of-plane mean-squared displacement; if the discrepancy exceeds the paper's claimed error bounds at moderate N, the two-point initialization is falsified.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that every oracle needed by the coupled-oscillator quantum algorithm can be implemented efficiently for a structured planar material, making the exponential space savings concrete rather than abstract. For graphene, the spring matrix has fixed sparsity d=3, and its translational symmetry lets the connectivity oracle be implemented with quantum arithmetic on unit-cell coordinates (row, column, sublattice) instead of memory lookups. Initial velocities are handled by a two-point discretization of the Maxwell-Boltzmann distribution that conserves the first two moments, so the kinetic energy is correct for large systems even while higher moments

Load-bearing premise

The central claim rests on the assumption that a thermal state in which every atom's velocity is either +σ or -σ with equal probability—chosen to match only the mean and variance of the Maxwell-Boltzmann distribution—captures the physics that heat transfer and rippling actually depend on; if higher velocity moments matter for those observables, the resource estimates no longer describe the intended simulation.

Editorial extensions

If this is right

  • A cm-square graphene sheet's vibrational dynamics can be encoded in roughly 160 logical qubits instead of about 180 petabytes of classical memory, a space saving that grows exponentially with atom count.
  • Heat-transfer simulations can track a propagating kinetic-energy hotspot through a binary search over subsets, giving a super-polynomial advantage for long-time dynamics.
  • Out-of-plane rippling can be extracted as a mean-squared displacement using the alternative encoding, at a high-order polynomial quantum speedup even though a classical dequantized algorithm exists.
  • The unit-cell oracle construction generalizes to other periodic planar materials, such as carbides or nitrides, by adjusting the sublattice basis and shift vectors.
  • Initial velocities sampled from a Maxwell-Boltzmann distribution can be loaded in polylogarithmic time using the discretized bucket approximation plus amplitude amplification, rather than exponential sampling.

Reading between the lines

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

  • Beyond the paper: because the two-bucket initialization matches only the first two moments of the velocity distribution, the heat-transfer and rippling predictions are only as trustworthy as the insensitivity of those observables to higher moments; a classical validation of this sensitivity would settle it before fault-tolerant hardware exists.
  • Beyond the paper: the roughly 160-logical-qubit estimate excludes fault-tolerance overhead, state-preparation ancillas, and measurement rounds, so the physical qubit count and runtime on an error-corrected machine could be orders of magnitude larger.
  • Beyond the paper: the two-bucket velocity loader could be tested directly as a classical random-velocity initialization; if it reproduces known phonon transport benchmarks, the quantum claim inherits that validation, and if not, the paper's general pseudorandom-function loader could supply more buckets at additional cost.
  • Beyond the paper: the harmonic approximation excludes phonon scattering and thermostats, so the two applications are proof-of-concept demonstrations; extending the framework with weak nonlinearities, as the paper suggests, would be the natural route to more realistic heat transport.
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 / 5 minor

Summary. The paper introduces Quantum Elastic Network Models (QENMs), applying the coupled-oscillator simulation algorithm of Babbush et al. (PRX 2023) to planar materials, with graphene as the worked example. The main contributions are an efficient connectivity oracle for a hexagonal lattice, a two-bucket discretization of the Maxwell–Boltzmann distribution for initial-velocity loading, resource estimates culminating in a claim of ~160 logical qubits for a 1 cm² graphene sheet, and two proposed applications: heat transfer and out-of-plane rippling. The paper is explicit about the distinction between long-time (heat transfer) and short-time (rippling) dynamics and about the dequantization results of Sakamoto and Fujii, and it candidly lists several limitations.

Significance. If the physical model and resource estimates were correct, the paper would be a valuable end-to-end demonstration of a fault-tolerant quantum algorithm for a structured, macroscopic harmonic system. The explicit connectivity oracle for a hexagonal lattice, the numerical scaling data for cond(B) and Tr(A⁺), and the honest treatment of dequantization are genuine strengths. However, the central physical identification of the simulated model with the ENM potential of Eq. (1) is incorrect, and the rippling application in particular is not derivable from the stated Hamiltonian. Because these issues affect the two headline applications and the resource claims, the current manuscript does not establish its main results.

major comments (4)
  1. [Sec. 4, Eqs. (37)–(40), and Sec. 4.5.3] The simulated Hamiltonian is not the Hessian of the ENM potential in Eq. (1). For U = (γ/2) Σ (|r_i−r_j|−r0_ij)², linearization about a flat lattice gives an orientation-dependent Hessian, not the scalar graph Laplacian F of Eq. (41) acting independently on x and y. For out-of-plane displacements, |r_i−r_j|−l0 ≈ (Δz)²/(2l0), so the potential is quartic in z and the z-Hessian is identically zero. Therefore the alternate encoding in Eq. (55) has P = 0 in the z-sector, F = 0 in Eq. (57), and the rippling relation ⟨M⟩ ∝ TA/κ_b in Eq. (84) cannot follow. A bending-rigidity or angle term is never defined. Application 2 is unsupported as stated.
  2. [Sec. 3.1.1, Eq. (24)] The two-bucket distribution B₂ᴰ matches only the first two moments of the Maxwell–Boltzmann distribution. The paper claims this is sufficient because it conserves kinetic energy, but it provides no error bound for time-evolved observables. The circuit prepares a single pseudorandom ±σ pattern, not a coherent superposition over Maxwell–Boltzmann samples; higher-order moments are badly wrong. For a purely harmonic system and quadratic observables, an ensemble average over random signs may preserve the covariance, but the paper uses a single realization and makes thermal and rippling claims. This requires either explicit concentration/error bounds or a restriction of the claims.
  3. [Sec. 5, qubit estimate] The headline estimate of ~160 logical qubits is not established. It counts 2 log₂ N + r + 2 and explicitly excludes state-preparation overhead. The circuits in Secs. 4.2–4.4 use additional ancilla registers for the dimension qubit, bucket assignment, velocity rotation, comparators, and controlled block-encoding operations. No bound on these overheads is given, and the assertion that they will not substantially increase the count is unsupported. Since the abstract’s central numerical claim depends on this estimate, it should be either proved with a full register accounting or removed.
  4. [Sec. 4.3 and Fig. 11] Fig. 11 reports cond(B) = O(√N). The initial-state preparation for the alternate displacement encoding Eq. (55) then costs Õ(√N log N) gates, which is exponential in n = log N. The text describes this as a “polynomial advantage over classical methods,” but it is not an efficient polylog(N) simulation. Because the rippling/MSD application relies on this encoding, the claim that the method enables efficient simulation of this observable is not supported. The comparison target should be stated explicitly, and the distinction between polylogarithmic and polynomial-in-N scaling should be made unambiguous.
minor comments (5)
  1. [Sec. 3.1.1, Eq. (24)] The subscripting in Eq. (24) is inconsistent: the probabilities are P₁, P₂, …, P_k while the representative velocities are written sv₀, sv₁, …. Please harmonize the indices.
  2. [Sec. 3.1.1] The phrase “load 2n samples” should be “2ⁿ samples”; the text alternates between these meanings and the distinction matters for the complexity discussion.
  3. [Sec. 4.1, Figs. 8–9] The neighbor-index register ℓ is described as a two-qubit register in state |0⟩+|1⟩+|2⟩. The circuits should make the encoding explicit (e.g., binary vs. one-hot) and explain how the value 3 is handled.
  4. [Sec. 4.1, padding] The padding strategy says the physical lattice has an odd number of rows and columns one less than a power of two, but the text later refers to an 8×8 padded lattice. Please clarify the relation between the physical atom count N, the padded row/column counts, and the resulting qubit count.
  5. [Sec. 4.5.2] The limitation paragraph on heat transfer correctly notes the difficulty of measuring exponentially small subset energies, but the sentence “only provides exponential advantage at very low temperature” would benefit from a quantitative statement of when K_V/E is polynomially large.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; core derivation is built on external [Bab+23]/[SF25] results and explicit oracles.

full rationale

The central derivation chain is: map the ENM harmonic potential (Eq. 1) through the Babbush coupled-oscillator encoding (Eqs. 6 and 11), construct a graphene connectivity oracle (Sec. 4.1), prepare Boltzmann-like velocities via the explicitly labeled two-bucket approximation B_2^D (Sec. 3.1.1), block-encode H (Eqs. 69-79), and measure subset kinetic energies or MSD (Sec. 4.5). The two-bucket initialization is an input approximation: matching the first two moments (Eq. 24) makes Corollary 3.2 a direct consequence of the definition, and it is used only to set the initial-state rotation angle, not to predict an independent observable. Heat-transfer and rippling outputs are time-evolved subset energies/MSD obtained from the external Hamiltonian-simulation framework; they are not refit to those outputs, nor is any benchmark defined by the paper's own fits. The only overlapping-author citations ([FW25], [KJN25], [Luo+24]) are passing references in algorithm lists or to a standard comparator primitive and are not load-bearing. The paper explicitly acknowledges physical-fidelity limitations, e.g. Sec. 4.5.2 states the harmonic approximation 'excludes fundamental anharmonic terms required for important phenomena like phonon scattering', and Sec. 4.5.3 notes thermostatting is 'currently beyond our current framework'; these are correctness/model-fidelity caveats rather than circular reductions. A separate scientific concern, not scored here, is that out-of-plane displacements have no quadratic restoring force in the central-force ENM of Eq. 1, making the rippling application physically unsupported; this is a validity issue, not a circularity.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central construction rests on (i) the harmonic ENM approximation, (ii) uniform mass and coupling for graphene, (iii) the specific periodic-padding model of the lattice, (iv) the numerically observed cond(B)=O(√N), and (v) the two-bucket thermal initialization. None of these is independently falsified in the paper; (v) is the most fragile. No new physical entities are introduced.

free parameters (3)
  • velocity truncation vmax
    Introduced in Sec. 3.1.1 to define the discretization intervals [−vmax, vmax]; chosen so that the Gaussian tails are negligible, but the error from truncation is not quantified.
  • number of velocity buckets k = 2
    Chosen so that the discrete distribution matches the first two moments only; higher-order moments of the Maxwell-Boltzmann distribution are not reproduced, which limits the physical fidelity of the initialization.
  • precision bits r for mass/spring oracles = ≈50
    Used in the ~160 logical qubit estimate (Sec. 5) to suppress errors in long-time dynamics; no rigorous error analysis ties r to the simulation accuracy target.
assumptions (5)
  • domain assumption The harmonic (quadratic) potential approximation of the ENM is accurate for the simulated time and displacements.
    This is the founding assumption of ENMs (Eq. (1)) and is also the stated limitation of the paper (Sec. 5: excludes anharmonicity, phonon scattering).
  • domain assumption All carbon atoms have identical mass and all bonds have identical spring constant κ.
    Assumed throughout Sec. 4 to simplify the oracles and the block encoding; the paper acknowledges that non-uniform couplings complicate the transformation (Eq. (33)).
  • ad hoc to paper The graphene lattice can be treated as a perfect rectangle with a power-of-two number of rows/columns after padding, and boundary effects are handled by dummy nodes.
    Sec. 4.1 assumes the filled rows/columns are one less than a power of two and adds padding; real CVD graphene sheets are not perfect rectangles of this form.
  • ad hoc to paper The condition number of the graphene incidence matrix scales as O(√N).
    Used in Sec. 4.3 to claim that the displacement-encoding preparation requires O~(√N log N) gates; only verified numerically for N≤5000 (Fig. 11), not proven analytically.
  • ad hoc to paper Initial velocities sampled from B_2^D(m,T) are an adequate proxy for a thermal ensemble.
    The 2-bucket discretization matches only the second moment; the paper provides no argument that this suffices for the heat-transfer or rippling observables beyond kinetic-energy estimates.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Quantum Elastic Network Models and their Application to Graphene." pith.science (2026). https://pith.science/paper/ORT25I3U

@misc{pith2026260105161,
  author       = {Pith},
  title        = {Pith review of: Quantum Elastic Network Models and their Application to Graphene},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ORT25I3U}},
  note         = {Machine review of arXiv:2601.05161}
}
abstract

Molecular dynamics simulations are a central computational methodology in materials design for relating atomic composition to mechanical properties. However, simulating materials with atomic-level resolution on a macroscopic scale is infeasible on current classical hardware, even when using the simplest elastic network models (ENMs) that represent molecular vibrations as a network of coupled oscillators. To address this issue, we introduce Quantum Elastic Network Models (QENMs) and utilize the quantum algorithm of Babbush et al. (PRX, 2023), which offers an exponential advantage when simulating systems of coupled oscillators. Here, we extend their algorithm in 2D systems and demonstrate how our method enables the efficient simulation of planar materials. As an example, we apply our algorithm to the task of simulating a 2D graphene sheet. We analyze the complexity for initial-state preparation, Hamiltonian simulation, and measurement of this material, and provide two real-world applications: heat transfer and the out-of-plane rippling effect. We estimate that an atomistic simulation of a graphene sheet on the centimeter scale, classically requiring hundreds of petabytes of memory and prohibitive runtimes, could be encoded and simulated with as few as $\sim 160$ logical qubits.

Figures

Figures reproduced from arXiv: 2601.05161 by the authors.

Figure 1
Figure 1. Illustration of a molecule and its corresponding ENM representation Panel (a) shows an illustration of a molecule, while panel (b) highlights the simplification in which atoms or groups of atoms are replaced by nodes connected (up to a desired cutoff distance) via springs. every atom into a mass, atoms are grouped (for example, it is common to label a carbon atom bonded to a functional group as α-carbon (C α)), and … view at source ↗
Figure 2
Figure 2. Overview of the algorithm introduced in [Bab+23]. The algorithm is divided into three main subroutines [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Illustration of sampling from Boltzmann distribution by sampling from D = 2 independent Gaussians Intuitively, each amplitude vi of every computational basis state |i⟩ indicates the velocity of atom i that is sampled from the Maxwell-Boltzmann probability distribution. Naively preparing a dense array of floats in the amplitude of the input state here would take exponential time. As mentioned, even classically sampli… view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Two-point approximation of the Maxwell-Boltzmann distribution. The probability mass is partitioned exactly at the median velocity to ensure equiprobable buckets (P0 = P1 = 0.5). The representative velocities v˜0 and v˜1 are chosen symmetrically around the mean (µ ± σ) …
Figure 5
Figure 5. Figure 5: Circuit for randomized bucket assigment: Given a randomly sampled θ = (s, r) ∈ {0, 1} n+1, and an n qubit quantum register |j⟩, the above construction computes (j · s) ⊕ r, where the dot product is in F2 1. Randomized Bucket Assignment We assign each node index j to a …
Figure 6
Figure 6. Figure 6: Velocity Amplitude Encoding Circuit. The abstract operation V (left) is implemented via controlled rotations (right). The flag qubit |bj ⟩ selects between rotation angles ϕ0 and ϕ1, mapping the normalized velocity v˜ into the ancilla amplitudes such that |ψbj ⟩ = q 1 −…
Figure 7
Figure 7. Figure 7: Graphene unit cell geometry and the padding strategy used in this work. (a) shows a graphene sheet. (b) shows the graphene lattice with unit cells and sublattices (orange and blue nodes) explicitly labeled. Solid nodes and dark lines correspond to physical carbon atoms…
Figure 8
Figure 8. Figure 8: Quantum circuit implementation of the Shift Initialization unitary (U⃗δ ). The circuits correspond to the four cases defined in [PITH_FULL_IMAGE:figures/full_fig_p020_8.png]
Figure 9
Figure 9. Figure 9: Circuit for computing neighbors by adding the relative shift vectors First, the unitary U⃗δ loads the relative shift vectors (δr, δc) inplace into the ℓ register, controlled by the source node parameters (r, s) and the neighbor index ℓ. Second, quantum adders perform t…
Figure 10
Figure 10. Figure 10: Scaling of the trace of the pseudo-inverse Laplacian [PITH_FULL_IMAGE:figures/full_fig_p026_10.png]
Figure 11
Figure 11. Figure 11: Condition number of the incidence matrix [PITH_FULL_IMAGE:figures/full_fig_p026_11.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

14 extracted references · 6 linked inside Pith

  1. [1]

    Empirical chemical pseudopotential theory of molecular and metallic bonding

    [Abe85] G. C. Abell. “Empirical chemical pseudopotential theory of molecular and metallic bonding”. In:Phys. Rev. B31 (10 May 1985), pp. 6184–6196. [AC25] Google Quantum AI and Collaborators. “Quantum error correction below the surface code threshold”. In:Nature638.8052 (2025), pp. 920–926. [Ala+25] Faisal Alam, Jan Lukas Bosse, Ieva ˇCepait˙ e, Adrian Ch...

    arXiv 1985

  2. [75]

    Nos´ e–Hoover chains: The canonical ensemble via continuous dynamics

    [MKT92] Glenn J Martyna, Michael L Klein, and Mark Tuckerman. “Nos´ e–Hoover chains: The canonical ensemble via continuous dynamics”. In:The Journal of chemical physics97.4 (1992), pp. 2635–2643. [MS13] Luca Monticelli and Emppu Salonen.Biomolecular simulations: methods and protocols. Vol

    1992

  3. [163]

    Scaling properties of flexible membranes from atomistic simulations: application to graphene

    [Los+09] JH Los, Mikhail I Katsnelson, OV Yazyev, KV Zakharchenko, and Annalisa Fasolino. “Scaling properties of flexible membranes from atomistic simulations: application to graphene”. In:Physical Review B—Condensed Matter and Materials Physics80.12 (2009), p. 121405. [Luo+24] Alessandro Luongo, Antonio Michele Miti, Varun Narasimhachar, and Adithya Sire...

    arXiv 2009

  4. [783]

    Empirical potential for hydrocarbons for use in simulating the chemical vapor deposition of diamond films

    [Bre90] Donald W. Brenner. “Empirical potential for hydrocarbons for use in simulating the chemical vapor deposition of diamond films”. In:Phys. Rev. B42 (15 Nov. 1990), pp. 9458–9471. [ ˇCep+25] Ieva ˇCepait˙ e, Niam Vaishnav, Leo Zhou, and Ashley Montanaro. “Quantum-Enhanced Optimization by Warm Starts”. In:arXiv preprint arXiv:2508.16309(2025). [Cer+21...

    arXiv 1990

  5. [1599]

    Quantum-enhanced markov chain monte carlo

    [Lay+23] David Layden, Guglielmo Mazzola, Ryan V Mishmash, Mario Motta, Pawel Wocjan, Jin-Sung Kim, and Sarah Sheldon. “Quantum-enhanced markov chain monte carlo”. In:Nature619.7969 (2023), pp. 282–287. [LC17] Guang Hao Low and Isaac L Chuang. “Optimal Hamiltonian simulation by quantum signal processing”. In:Physical review letters118.1 (2017), p. 010501....

    2023

  6. [1905]

    Exponential Quantum Advantage for Simulating Open Classical Systems

    [VYM25] Agi Villanyi, Yariv Yanay, and Ari Mizel. “Exponential Quantum Advantage for Simulating Open Classical Systems”. In:arXiv preprint arXiv:2503.11483(2025). [Wei+13] Yujie Wei, Baoling Wang, Jiangtao Wu, Ronggui Yang, and Martin L Dunn. “Bending rigidity and Gaussian bending stiffness of single-layered graphene”. In:Nano letters 13.1 (2013), pp. 26–...

    arXiv 2025

  7. [1997]

    Quantum algorithm for simulating the wave equation

    [CJO19] Pedro CS Costa, Stephen Jordan, and Aaron Ostrander. “Quantum algorithm for simulating the wave equation”. In:Physical Review A99.1 (2019), p. 012323. [CKS17] Andrew M Childs, Robin Kothari, and Rolando D Somma. “Quantum algorithm for systems of linear equations with exponentially improved dependence on precision”. In: SIAM Journal on Computing46....

    2019

  8. [2004]

    Chemistry beyond the scale of exact diagonalization on a quantum-centric supercomputer

    [Rob+25] Javier Robledo-Moreno, Mario Motta, Holger Haas, Ali Javadi-Abhari, Petar Jurcevic, William Kirby, Simon Martiel, Kunal Sharma, Sandeep Sharma, Tomonori Shirakawa, et al. “Chemistry beyond the scale of exact diagonalization on a quantum-centric supercomputer”. In:Science Advances11.25 (2025), eadu9991. [San+19] Yuval R Sanders, Guang Hao Low, Art...

    arXiv 2025

Show all 14 references
  1. [2013]

    Simulating Time Dependent and Nonlin- ear Classical Oscillators through Nonlinear Schr \

    [MW25] Abhinav Muraleedharan and Nathan Wiebe. “Simulating Time Dependent and Nonlin- ear Classical Oscillators through Nonlinear Schr \” odingerization”. In:arXiv preprint arXiv:2505.17170(2025). [OBr+22] Thomas E O’Brien, Michael Streif, Nicholas C Rubin, Raffaele Santagati,...

    2025 arXiv

  2. [2023]

    Quantum-enhanced Markov chain Monte Carlo for systems larger than a quantum computer

    36 [FW25] Stuart Ferguson and Petros Wallden. “Quantum-enhanced Markov chain Monte Carlo for systems larger than a quantum computer”. In:Physical Review Research7.1 (2025), p. 013231. [GM15] Elaheh K Goharshadi and Sayyed Jalil Mahdizadeh. “Thermal conductivity and heat transp...

    2025

  3. [2767]

    A new quantum ripple-carry addition circuit

    [Cuc+04] Steven A Cuccaro, Thomas G Draper, Samuel A Kutin, and David Petrie Moulton. “A new quantum ripple-carry addition circuit”. In:arXiv preprint quant-ph/0410184 (2004). [DM11] Jacob D Durrant and J Andrew McCammon. “Molecular dynamics simulations and drug discovery”. In...

    2004 arXiv

  4. [3090]

    Normal modes and essential dynamics

    [HD08] Steven Hayward and Bert L De Groot. “Normal modes and essential dynamics”. In: Molecular Modeling of Proteins(2008), pp. 89–106. [HD18] Scott A Hollingsworth and Ron O Dror. “Molecular dynamics simulation for all”. In: Neuron99.6 (2018), pp. 1129–1143. [HFX11] Feng Hao,...

    2008

  5. [3720]

    Protein elastic network models and the ranges of cooperativity

    [YSJ09] Lei Yang, Guang Song, and Robert L Jernigan. “Protein elastic network models and the ranges of cooperativity”. In:Proceedings of the National Academy of Sciences106.30 (2009), pp. 12347–12352. 39 [Zha+16] Jinjun Zhang, Bonsung Koo, Nithya Subramanian, Yingtao Liu, and ...

    2009

  6. [3899]

    Large amplitude elastic motions in proteins from a single-parameter, atomic analysis

    [Tir96] Monique M Tirion. “Large amplitude elastic motions in proteins from a single-parameter, atomic analysis”. In:Physical review letters77.9 (1996), p

    1996

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.