Pith. sign in

REVIEW 2 major objections 6 minor 66 references

Non-reciprocal turn-towards torques can assemble active particles into clusters whose symmetry dictates whether they sit still, translate, or rotate—so the particle number and torque strength become programing knobs for micro-machine functi

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

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

T0 review · deepseek-v4-flash

2026-08-03 02:32 UTC pith:CXT3RFV4

load-bearing objection Solid simulation study with a convincing central mechanism; the 'N dictates function' claim needs stronger evidence but does not undermine the core result. the 2 major comments →

arxiv 2607.29651 v1 pith:CXT3RFV4 submitted 2026-07-31 cond-mat.soft cond-mat.stat-mech

Non-reciprocal torques guide self-assembly of active particles into clusters with controllable function

Till Welker , Yukino Fujiya , Holger Stark This is my paper
classification cond-mat.soft cond-mat.stat-mech
keywords active matterself-assemblycolloidal machinesnon-reciprocal torquesturn-towards torquerun-and-tumble motionstochastic resettingsymmetry and dynamics
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Dense groups of self-propelled particles that turn toward their neighbours can be held together without any attractive force. Because every particle's orientation is tied to the local arrangement of neighbours, the finished cluster's shape determines the pattern of propulsion directions, and symmetry then fixes the cluster's motion: two mirror axes mean a static cluster, one mirror axis permits translation along it, pure point symmetry permits rotation, and no symmetry should give spiralling. For clusters of up to five particles the particle number alone selects the stable shape, so one parameter programs the machine's function. For six particles the system is multi-stable—a static triangle, a translating chevron, and a rotating parallelogram—and the torque strength biases which function dominates; the chevron can also tumble, giving run-and-tumble motion. Because the interaction can be switched off, briefly turning off activity implements a stochastic reset that cuts slow assembly pathways and speeds up assembly by a factor of about 3.5.

Core claim

The paper establishes that a short-range torque turning each particle toward its neighbours creates self-assembled clusters whose function is controlled by symmetry. The torque Γ_ij = Γ0 (n_i × r_hat_ij) couples the orientation field to particle positions; summing the single-particle equations of motion then shows that the cluster's propulsion velocity is v0 times the mean orientation, and its angular velocity is determined by the distribution of positions and orientations. Applying the Neumann-Minnigerode-Curie principle—a property of an object is at least as symmetric as the object itself—to the resulting combined symmetry gives the allowed motions: clusters with two symmetry axes are stat

What carries the argument

The central object is the turn-towards torque, Γ_ij = Γ0 (n_i × r_hat_ij), active only between nearest neighbours; it makes a particle rotate its propulsion direction toward the mean direction of its neighbours. That torque couples position and orientation, so the cluster's configurational symmetry constrains the orientational symmetry and, through the Neumann-Minnigerode-Curie principle, the cluster's allowed translational and rotational velocities (Eqs. 2–3). The numerical state classification additionally relies on the sorted neighbour-number list, which the paper assumes uniquely identifies a configuration; and the run-and-tumble analysis rests on exponentially distributed run times and

Load-bearing premise

The load-bearing premise is that the sorted neighbour-number list uniquely identifies the cluster configuration, together with the claim that each particle number up to five has exactly one stable state; if two distinct geometries share a neighbour-number list, or a missing stable state exists, the measured occupation probabilities and the conclusion that particle number dictates function could be misassigned.

What would settle it

Enumerate all connected rigid clusters of six disks under the turn-towards torque, compute each cluster's sorted neighbour-number list, and simulate its motion in isolation: if two distinct geometries share the same list, or if a listed 'stable' geometry does not show the symmetry-predicted motion (triangle diffusive, chevron translating, parallelogram rotating), the central claim is falsified.

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

Share X Bluesky LinkedIn Reddit HN

If this is right

  • For N ≤ 5, particle number alone selects the stable cluster configuration and therefore the cluster's function, from static to translating to rotating.
  • For N = 6, the torque-velocity ratio Γ0/v0 becomes a second control knob: increasing it biases the population from static triangles toward translating chevrons, and beyond a threshold the chevron performs run-and-tumble motion.
  • Because the turn-towards interaction is switchable, clusters assemble only while activity is on, enabling just-in-time assembly and disassembly.
  • Briefly switching off propulsion and torque implements stochastic resetting, which in the four-particle case cuts the mean assembly time by about a factor of 3.5 and the spread of assembly times by a factor of about 6.7.

Where Pith is reading between the lines

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

  • Inference: the symmetry-to-motion mapping should hold for any active system in which orientations are locked to local neighbour geometry, so the same programing scheme may transfer to other torque-generation mechanisms beyond the specific rule simulated here.
  • Inference: because the chevron's run-and-tumble rate and tumbling-angle distribution are cleanly separated from orientational diffusion, tuning torque strength or noise could give independent control over persistence length, not just over state occupation.
  • Inference: the finite-time reset implementation leaves an obvious optimisation margin; jointly optimising reset rate r and reset duration τ should reduce mean assembly time further than the demonstrated proof-of-concept values r=0.1, τ=5.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 6 minor

Summary. The paper studies two-dimensional active Brownian particles interacting through a non-reciprocal 'turn-towards' torque (Eq. 1) plus repulsive WCA forces, with no attraction. It derives exact-like expressions for the cluster center-of-mass velocity and angular velocity (Eqs. 2 and 3), uses symmetry (Neumann–Minnigerode–Curie) to classify allowed cluster functions (static, translating, rotating), and reports that small clusters have one stable configuration per particle number while six-particle clusters exhibit three metastable configurations (triangle, chevron, parallelogram) whose occupation probabilities and transition fluxes are tunable via the torque-to-velocity ratio Γ0/v0. The chevron is shown to perform run-and-tumble motion. For four-particle assembly, two pathways are identified and stochastic resetting, implemented by temporarily switching off activity, is shown to reduce the mean assembly time by a factor of 3.5. The paper claims that particle number and torque strength provide two complementary control knobs for programming colloidal micro-machine function.

Significance. If the central claims hold, the paper offers a clean design principle: non-reciprocal torques couple orientation to cluster geometry, so symmetry alone determines whether a cluster translates, rotates, or remains static. The paper has several concrete strengths: the threshold Γ0/v0 ≈ 2/3 follows from a simple timescale argument and is quantitatively checked in Fig. 4c; the seven-particle MSD model in Fig. 3 is parameter-free and matches simulation; the resetting speed-up is demonstrated with large simulation counts (65,000 and 10,000 runs); and the code is openly available. The main weakness is that the state classifier underlying the central 'particle number dictates function' and 'tunable multi-stability' conclusions rests on unproven uniqueness and exhaustiveness claims.

major comments (2)
  1. [Methods IV B and Fig. 4] The state classifier b is the sorted degree sequence of the contact graph. This invariant is not injective on planar configurations: for N=4, the path P4 and the 'T' graph both give b=(1,1,2,2); the six-particle sequences c2 and c3 in Fig. 7 also correspond to many possible simple graphs. The paper only states that 'in the cases considered here' uniqueness holds, but no proof or enumeration is supplied. Since the occupation probabilities P_i, fluxes J_ij (Fig. 4c,d), the tumbling rate (Fig. 5b), and the transition network all rely on this assignment, any transient or undiscovered configuration sharing a b with c1, c2, or c3 is silently merged into a stable state. Please validate the classifier against exact particle positions (e.g., canonical graph labeling or geometric template matching) and quantify how often distinct geometries share the same b in the simulations.
  2. [Sec. II C] The statement 'for clusters up to five particles, there is only one stable state for each particle number' is introduced without a definition of stability and without supporting evidence. No random-initial-condition sweep, no parameter study, and no exhaustive enumeration is provided for N≤5; Fig. 2 shows only selected examples. This claim is the basis of the abstract's 'particle number uniquely determines the stable configuration and function' and of the two-knob control narrative. Please make 'stable state' operational (e.g., lifetime threshold or basin definition) and demonstrate exhaustiveness for N=2–5 over the parameter range used. Otherwise the claim is indistinguishable from 'only one state was observed in the runs shown.'
minor comments (6)
  1. [Eq. (3)] The derivation text says 'taking the cross product of r_i with Eq. (4)', but the algebra uses r_i − r_c. Please make this explicit to avoid confusion.
  2. [Methods IV A / Fig. 6] For the 65,000 assembly runs, the initial condition 'one particle randomly placed in each quadrant' should specify the distribution (uniform?) and whether the quadrant assignment is random per run.
  3. [Fig. 4c] The occupation probabilities in Fig. 4c are shown without error bars, while Fig. 4d has shaded Poisson errors. Please add analogous uncertainty estimates to Fig. 4c.
  4. [Fig. 5b] The exponential fit p_r,fit = c e^{-ct} with c=0.235 is reported without a goodness-of-fit measure or confidence interval. A chi-squared value or similar would support the claim of exponential run time distribution.
  5. [Methods IV B / Fig. 7] The notation ilde b and b is confusingly similar; both are called 'neighbour numbers'. Please use distinct symbols (e.g., d_i for unsorted degrees and s for sorted list) throughout.
  6. [Fig. 2] The caption and text do not state the particle numbers for the clusters in Fig. 2a–c, which makes it hard to connect the examples to the 'one stable state per N≤5' claim. Please add the N values to the caption or text.

Circularity Check

0 steps flagged

No significant circularity: central derivations are self-contained; flagged state-classification concerns are verification issues, not circular reductions.

full rationale

The paper's central derivation chain is self-contained. Equations (2) and (3) are derived by summing the Langevin equations of motion (4)–(5), and the symmetry classification follows by applying the stated Neumann-Minnigerode-Curie transformations to those equations; the symmetry restrictions are not inputs loaded from the conclusion. The clustering-onset estimate Γ0/v0 ~ γ_r/R = 2/3 is a parameter-free timescale argument from the model parameters, checked against the simulated onset rather than fitted to it. The seven-particle MSD model uses the known active Brownian formula with v_c = v0/7, D_t,c = D_t/7, and τ_c = 1, all evaluated from the equations of motion, with no fitted parameters. The run-and-tumble analysis measures c = 0.235 and <cos Δφ> = -0.721 from simulations but uses them only to decompose the measured orientation correlation function; it does not generate a predicted quantity from fitted inputs. The sorted-neighbour-list classifier in Methods IV B is an operational coarse-graining of configurations; the two claims flagged in review — that the sorted list uniquely determines the configuration and that there is only one stable state for each particle number up to five — are unproven empirical/algorithmic assertions and therefore verification risks, but they are not circular: no conclusion is defined into the classifier, and no fitted parameter is renamed as a prediction. The self-citations [38,40] are background references for related torque models and are not load-bearing for the paper's main results.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 0 invented entities

The paper introduces no new particles, forces, or conserved quantities; the turn-toward torque interaction is taken from prior literature. The model parameters Γ0, v0, R, ε, Dt, Dr define the simulation regime rather than being fitted to a target result. The only data-fitted numbers are the descriptive tumbling rate and the hand-chosen reset rate/duration for a proof-of-concept demonstration. The main burden is the domain assumption that this torque model maps onto experimental Janus particles and that the symmetry principle carries over to active steady states.

free parameters (3)
  • tumbling rate c = 0.235
    Fitted to the exponential run-time distribution in Fig. 5b; used post hoc to decompose orientation decorrelation, not to set the model.
  • reset rate r = 0.1
    Chosen by hand for the proof-of-concept resetting run; not optimized and not fitted to data.
  • reset duration τ = 5
    Chosen by hand so that particles diffuse out of interaction range during a reset; not optimized.
axioms (4)
  • domain assumption Overdamped Langevin equations (4)-(5) with delta-correlated white noise describe particle motion.
    Standard model for colloidal active particles; not derived in the paper.
  • domain assumption Turn-towards torque (1) with cutoff R=1.5σ and WCA repulsion captures the relevant experimental interactions.
    Borrowed from prior work (Refs. [34-40,51]); the paper studies consequences rather than deriving the torque law.
  • domain assumption Neumann-Minnigerode-Curie symmetry principle applies to the non-equilibrium steady states of active clusters.
    Used in Sec. II B to translate cluster symmetry into allowed motion; no proof for active NESS states is given.
  • ad hoc to paper The sorted neighbour list uniquely and completely identifies stable cluster states for N≤7.
    Assumed in Sec. IV B; underpins occupation probabilities and flux measurements; completeness and uniqueness are not proven.

pith-pipeline@v1.3.0-daily-deepseek · 14155 in / 12302 out tokens · 137663 ms · 2026-08-03T02:32:33.941469+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Non-reciprocal torques guide self-assembly of active particles into clusters with controllable function." pith.science (2026). https://pith.science/paper/CXT3RFV4

@misc{pith2026260729651,
  author       = {Pith},
  title        = {Pith review of: Non-reciprocal torques guide self-assembly of active particles into clusters with controllable function},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CXT3RFV4}},
  note         = {Machine review of arXiv:2607.29651}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Self-assembly of constituents determines structure formation in the microscopic world. Attractive forces can assemble active particles into colloidal machines, but they do not fix the particles' orientations, which limits control over the machine's function. We demonstrate that non-reciprocal turn-towards torques not only assemble active particles into clusters, without requiring attractive forces, but also link particle orientations to the cluster configuration. Symmetry then dictates whether the cluster is static, rotates, or translates. In small systems, the particle number uniquely determines the stable configuration and function. In larger systems, there are multiple stable configurations with distinct functions, and tuning the torque strength allows us to bias towards the desired function, such as a run-and-tumble motion. Because the interactions driving assembly can be switched on and off, the clusters self-assemble when needed. For such a "just-in-time" self-assembly to be practical, fast assembly is necessary. We show that stochastic resetting, implemented by briefly turning off propulsion and torque, significantly speeds up self-assembly by avoiding slow pathways. Together, our findings demonstrate that non-reciprocal torques can rapidly assemble active particles into colloidal micromachines with controllable function.

Figures

Figures reproduced from arXiv: 2607.29651 by Holger Stark, Till Welker, Yukino Fujiya.

Figure 1
Figure 1. Figure 1: FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: a). The outer particles are all oriented towards the cen￾tre and their propulsions cancel by symmetry. In contrast, the 10 1 100 101 102 Lag time t 100 101 102 103 MSD r2 c a b FIG. 3. Free inner particle. a) Cluster of seven particles. The turn-towards torque orients the outer particles inward. In contrast, the torque on the inner particle vanishes because of the symmetry of the environment, and it can di… view at source ↗
Figure 4
Figure 4. Figure 4: c). For very small torques, there are no clusters. But as the torque-velocity ratio surpasses Γ0/v0 ≈ 2/3, triangle and parallelogram clusters form. Increasing Γ0/v0 increases the probability of a chevron cluster until it completely dominates the population around Γ0/v0 ∼ 5. To understand the onset of clustering, we consider the rele￾vant timescales in our system: (1) the time to propel over the interactio… view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6 [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7 [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

66 extracted references · 3 linked inside Pith

  1. [1]

    As illustrated in Fig

    Symmetry dictates function in small clusters Turn-towards torques relate symmetries of the positions and orientations to each other. As illustrated in Fig. 1c), parti- cles turn towards the mean distance vector in their neighbour- hood⟨ ˆrij⟩j∈Si. Consequently, the preferred orientation fol- lows the local symmetry of the particle neighbourhood. If all lo...

  2. [2]

    However, if a particle has a symmetric environment (for example in the bulk), the torques acting from the neighbours cancel, and the particle orienta- tion diffuses freely

    Bulk particles rotate freely The connection between the symmetry of particle positions and symmetry of orientations requires all particle orientations to be linked to the positions. However, if a particle has a symmetric environment (for example in the bulk), the torques acting from the neighbours cancel, and the particle orienta- tion diffuses freely. Le...

  3. [3]

    Occupation Probabilities and Reconfigurations The occupation probability of the system being in the trian- gle (i= 1), chevron (i= 2), or parallelogram (i= 3) state depends on the ratio of torque and velocityΓ 0/v0 as shown in Fig. 4c). For very small torques, there are no clusters. But as the torque-velocity ratio surpassesΓ 0/v0 ≈2/3, triangle and paral...

  4. [4]

    leader particle

    Run-and-Tumble motion The six-particle system has three different metastable mor- phologies (’mesostates’). Because all particles are identi- cal, there are multiple sub-configurations (’microstates’) re- sulting in the same mesostate. And indeed, for the translat- ing chevron cluster we observe transitions between those mi- crostates. These reconfigurati...

  5. [5]

    chemical notation

    Assembly pathways and timescales Let us follow the assembly pathways shown in Fig. 6a). Starting from four free particles, the only possible assembly event is the collision and merger of two particles, resulting in one two-particle cluster and two remaining free particles. But now there are two possible paths to proceed. (1) The re- maining free particles...

  6. [6]

    just-in-time

    Expedite self-assembly with stochastic resetting Kinetic traps pose a major challenge for efficient self- assembly. In our system, pathway (1) significantly slows down the assembly process. In this section, we will show how a physical implementation of stochastic resetting can signifi- cantly expedite assembly by escaping the slow pathway. The assembly of...

  7. [7]

    G. M. Whitesides and B. Grzybowski, Self-assembly at all scales, Science295, 2418 (2002)

    2002

  8. [8]

    Whitelam and R

    S. Whitelam and R. L. Jack, The statistical mechanics of dy- namic pathways to self-assembly, Annual review of physical chemistry66, 143 (2015)

    2015

  9. [9]

    Zeravcic, V

    Z. Zeravcic, V . Manoharan, and M. Brenner, Colloquium: To- ward living matter with colloidal particles, Reviews of Modern Physics89(2017)

    2017

  10. [10]

    Nguyen, Y

    M. Nguyen, Y . Qiu, and S. Vaikuntanathan, Organization and self-assembly away from equilibrium: Toward thermodynamic design principles, Annual Review of Condensed Matter Physics 12, 273 (2021)

    2021

  11. [11]

    F. Li, D. P. Josephson, and A. Stein, Colloidal assembly: the road from particles to colloidal molecules and crystals, Ange- wandte Chemie International Edition50, 360 (2011)

    2011

  12. [12]

    Moisant, H

    P. Moisant, H. Neeman, and A. Zlotnick, Exploring the paths of (virus) assembly, Biophysical journal99, 1350 (2010)

    2010

  13. [13]

    J. D. Perlmutter and M. F. Hagan, Mechanisms of virus assem- bly, Annual review of physical chemistry66, 217 (2015)

    2015

  14. [14]

    T. Sych, Y . M ´ely, and W. R ¨omer, Lipid self-assembly and lectin-induced reorganization of the plasma membrane, Philo- sophical Transactions of the Royal Society B: Biological Sci- ences373, 20170117 (2018)

    2018

  15. [15]

    K. J. Bishop, S. L. Biswal, and B. Bharti, Active colloids as models, materials, and machines, Annual Review of Chemical and Biomolecular Engineering14, 1 (2023)

    2023

  16. [16]

    F. M. Gartner, I. R. Graf, and E. Frey, The time complexity of self-assembly, Proceedings of the National Academy of Sci- ences119, e2116373119 (2022)

    2022

  17. [17]

    Sacanna, W

    S. Sacanna, W. T. Irvine, P. M. Chaikin, and D. J. Pine, Lock and key colloids, Nature464, 575 (2010)

    2010

  18. [18]

    Hormoz and M

    S. Hormoz and M. P. Brenner, Design principles for self- assembly with short-range interactions, Proceedings of the Na- tional Academy of Sciences108, 5193 (2011)

    2011

  19. [19]

    Murugan, Z

    A. Murugan, Z. Zeravcic, M. P. Brenner, and S. Leibler, Mul- tifarious assembly mixtures: Systems allowing retrieval of di- verse stored structures, Proceedings of the National Academy of Sciences112, 54 (2015)

    2015

  20. [20]

    T. K. Haxton and S. Whitelam, Do hierarchical structures as- semble best via hierarchical pathways?, Soft Matter9, 6851 (2013)

    2013

  21. [21]

    E. M. King, C. X. Du, Q.-Z. Zhu, S. S. Schoenholz, and M. P. Brenner, Programming patchy particles for materials assembly design, Proceedings of the National Academy of Sciences121, e2311891121 (2024)

    2024

  22. [22]

    Brandm ¨uller, An extension of the neumann–minnigerode– curie principle, inSymmetry(Elsevier, 1986) pp

    J. Brandm ¨uller, An extension of the neumann–minnigerode– curie principle, inSymmetry(Elsevier, 1986) pp. 97–100

    1986

  23. [23]

    A. M. Brooks, S. Sabrina, and K. J. Bishop, Shape-directed dynamics of active colloids powered by induced-charge elec- trophoresis, Proceedings of the national academy of sciences 115, E1090 (2018)

    2018

  24. [24]

    Aubret, Q

    A. Aubret, Q. Martinet, and J. Palacci, Metamachines of pluripotent colloids, Nature communications12, 6398 (2021)

    2021

  25. [25]

    L ¨owen, Active colloidal molecules, Europhysics Letters 121, 58001 (2018)

    H. L ¨owen, Active colloidal molecules, Europhysics Letters 121, 58001 (2018)

    2018

  26. [26]

    W. Gao, A. Pei, X. Feng, C. Hennessy, and J. Wang, Organized self-assembly of janus micromotors with hydrophobic hemi- spheres, Journal of the American Chemical Society135, 998 (2013)

    2013

  27. [27]

    Kaiser, K

    A. Kaiser, K. Popowa, and H. L ¨owen, Active dipole clusters: From helical motion to fission, Physical Review E92, 012301 (2015)

    2015

  28. [28]

    Guzm ´an-Lastra, A

    F. Guzm ´an-Lastra, A. Kaiser, and H. L¨owen, Fission and fusion scenarios for magnetic microswimmer clusters, Nature commu- nications7, 13519 (2016)

    2016

  29. [29]

    Zhang, J

    J. Zhang, J. Yan, and S. Granick, Directed self-assembly path- ways of active colloidal clusters, Angewandte Chemie128, 5252 (2016)

    2016

  30. [30]

    J. N. Johnson, A. Nourhani, R. Peralta, C. McDonald, B. Thiesing, C. J. Mann, P. E. Lammert, and J. G. Gibbs, Dy- namic stabilization of janus sphere trans-dimers, Physical Re- view E95, 042609 (2017)

    2017

  31. [31]

    Z. Shen, A. W ¨urger, and J. S. Lintuvuori, Hydrodynamic self- assembly of active colloids: chiral spinners and dynamic crys- tals, Soft matter15, 1508 (2019)

    2019

  32. [32]

    A. G. Subramaniam, M. Kumar, S. Thutupalli, and R. Singh, Rigid flocks, undulatory gaits, and chiral foldamers in a chem- ically active polymer, New Journal of Physics26, 083009 (2024)

    2024

  33. [33]

    Rosenberg and H

    M. Rosenberg and H. L ¨owen, Windmilling clusters of active quadrupoles, arXiv preprint arXiv:2509.26369 (2025)

    arXiv 2025

  34. [34]

    M. S. D. Wykes, J. Palacci, T. Adachi, L. Ristroph, X. Zhong, M. D. Ward, J. Zhang, and M. J. Shelley, Dynamic self- assembly of microscale rotors and swimmers, Soft matter12, 4584 (2016)

    2016

  35. [35]

    Soto and R

    R. Soto and R. Golestanian, Self-assembly of catalytically active colloidal molecules: tailoring activity through surface chemistry, Physical review letters112, 068301 (2014)

    2014

  36. [36]

    Soto and R

    R. Soto and R. Golestanian, Self-assembly of active col- loidal molecules with dynamic function, Physical Review E91, 052304 (2015)

    2015

  37. [37]

    Varma, T

    A. Varma, T. D. Montenegro-Johnson, and S. Michelin, Clustering-induced self-propulsion of isotropic autophoretic particles, Soft matter14, 7155 (2018)

    2018

  38. [38]

    Schmidt, B

    F. Schmidt, B. Liebchen, H. L ¨owen, and G. V olpe, Light- controlled assembly of active colloidal molecules, The Journal of chemical physics150(2019)

    2019

  39. [39]

    R. F. Ismagilov, A. Schwartz, N. Bowden, and G. M. White- sides, Autonomous movement and self-assembly, Angewandte Chemie International Edition41, 652 (2002)

    2002

  40. [40]

    Smeets, R

    B. Smeets, R. Alert, J. Pe ˇsek, I. Pagonabarraga, H. Ramon, and R. Vincent, Emergent structures and dynamics of cell colonies by contact inhibition of locomotion, Proceedings of the Na- tional Academy of Sciences113, 14621 (2016)

    2016

  41. [41]

    Nilsson and G

    S. Nilsson and G. V olpe, Metastable clusters and channels formed by active particles with aligning interactions, New Jour- nal of Physics19, 115008 (2017)

    2017

  42. [42]

    Zhang, R

    J. Zhang, R. Alert, J. Yan, N. S. Wingreen, and S. Granick, Active phase separation by turning towards regions of higher density, Nature Physics17, 961 (2021)

    2021

  43. [43]

    S. Das, M. Ciarchi, Z. Zhou, J. Yan, J. Zhang, and R. Alert, Flocking by turning away, Physical Review X14, 031008 (2024)

    2024

  44. [44]

    Kne ˇzevi´c, T

    M. Kne ˇzevi´c, T. Welker, and H. Stark, Collective motion of active particles exhibiting non-reciprocal orientational interac- tions, Scientific reports12, 19437 (2022)

    2022

  45. [45]

    Shea and H

    J. Shea and H. Stark, Emergent collective behavior of cohe- sive, aligning particles, The European Physical Journal E48, 22 (2025)

    2025

  46. [46]

    Welker and R

    T. Welker and R. Alert, Lattice-dependent orientational order in active crystals, Soft Matter21, 7228 (2025)

    2025

  47. [47]

    Q.-Z. Zhu, C. X. Du, E. M. King, and M. P. Brenner, Proof- 10 reading mechanism for colloidal self-assembly, Physical Re- view Research6, L042057 (2024)

    2024

  48. [48]

    J. F. Schubert, S. F. Navas, and S. H. Klapp, Self-assembly and time-dependent control of active and passive triblock janus col- loids, arXiv preprint arXiv:2504.20764 (2025)

    Pith/arXiv arXiv 2025

  49. [49]

    Dopierała, L

    D. Dopierała, L. Cocconi, R. L. Jack, and A. Souslov, Odd path- ways speed up self-assembly, arXiv preprint arXiv:2604.22408 (2026)

    Pith/arXiv arXiv 2026

  50. [50]

    Faran and G

    M. Faran and G. Bisker, Nonequilibrium self-assembly control by the stochastic landscape method, Journal of Chemical Infor- mation and Modeling65, 4067 (2025)

    2025

  51. [51]

    Z. M. Sherman and J. W. Swan, Dynamic, directed self- assembly of nanoparticles via toggled interactions, ACS nano 10, 5260 (2016)

    2016

  52. [52]

    Liang, M

    Z. Liang, M. X. Lim, Q.-Z. Zhu, F. Mottes, J. Z. Kim, L. Gut- tieres, C. Smart, T. Pearson, C. X. Du, M. Brenner,et al., Mag- netic decoupling as a proofreading strategy for high-yield, time- efficient microscale self-assembly, Proceedings of the National Academy of Sciences122, e2502361122 (2025)

    2025

  53. [53]

    M. R. Evans and S. N. Majumdar, Diffusion with optimal reset- ting, Journal of Physics A: Mathematical and Theoretical44, 435001 (2011)

    2011

  54. [54]

    M. R. Evans, S. N. Majumdar, and G. Schehr, Stochastic reset- ting and applications, Journal of Physics A: Mathematical and Theoretical53, 193001 (2020)

    2020

  55. [55]

    Reuveni, Optimal stochastic restart renders fluctuations in first passage times universal, Physical review letters116, 170601 (2016)

    S. Reuveni, Optimal stochastic restart renders fluctuations in first passage times universal, Physical review letters116, 170601 (2016)

    2016

  56. [56]

    A. Pal, S. Kostinski, and S. Reuveni, The inspection paradox in stochastic resetting, Journal of Physics A: Mathematical and Theoretical55, 021001 (2022)

    2022

  57. [57]

    J. Yan, M. Han, J. Zhang, C. Xu, E. Luijten, and S. Granick, Reconfiguring active particles by electrostatic imbalance, Na- ture materials15, 1095 (2016)

    2016

  58. [58]

    Z ¨ottl and H

    A. Z ¨ottl and H. Stark, Emergent behavior in active colloids, Journal of Physics: Condensed Matter28, 253001 (2016)

    2016

  59. [59]

    R. W. Perry, M. C. Holmes-Cerfon, M. P. Brenner, and V . N. Manoharan, Two-dimensional clusters of colloidal spheres: Ground states, excited states, and structural rearrangements, Physical review letters114, 228301 (2015)

    2015

  60. [60]

    Datta, C

    A. Datta, C. Beta, and R. Großmann, Random walks of inter- mittently self-propelled particles, Physical Review Research6, 043281 (2024)

    2024

  61. [61]

    H. C. Berg and D. A. Brown, Chemotaxis in escherichia coli analysed by three-dimensional tracking, nature239, 500 (1972)

    1972

  62. [62]

    Baconnier, D

    P. Baconnier, D. Shohat, C. H. L ´opez, C. Coulais, V . D´emery, G. D ¨uring, and O. Dauchot, Selective and collective actuation in active solids, Nature Physics18, 1234 (2022)

    2022

  63. [63]

    Baconnier, O

    P. Baconnier, O. Dauchot, V . D ´emery, G. D ¨uring, S. Henkes, C. Huepe, and A. Shee, Self-aligning polar active matter, Re- views of Modern Physics97, 015007 (2025)

    2025

  64. [64]

    Kruse, J.-P

    K. Kruse, J.-P. Eckmann, and W. C. Poon, Active self- disassembly enhances the yield of self-assembled structures, arXiv preprint arXiv:2405.07239 (2024)

    Pith/arXiv arXiv 2024

  65. [65]

    K. S. Olsen, D. Gupta, F. Mori, and S. Krishnamurthy, Thermo- dynamic cost of finite-time stochastic resetting, Physical Re- view Research6, 033343 (2024)

    2024

  66. [66]

    Tal-Friedman, Y

    O. Tal-Friedman, Y . Roichman, and S. Reuveni, Diffusion with partial resetting, Physical Review E106, 054116 (2022). Data availability -The code required to repro- duce the results of this paper is openly available at https://github.com/tillwelker/Non-reciprocal-torques-guide- self-assembly-of-active-particles. Acknowledgements -We thank Juri Schubert, P...

    2022