{"id":"b4e98dca-9637-46e1-9d0a-c0e58360c41d","arxiv_id":"2411.15544","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A generalized Ito calculus connects individual run-and-tumble paths to their probability equations, yielding new pathwise derivations for resetting, sticky boundaries, entropy, and global-reset populations.","lead":"A mathematician develops a path-by-path calculus for run-and-tumble particles, which move at constant speed and randomly flip direction. The method turns previously assumed probability equations into derivations and reveals that a population-wide reset couples otherwise independent particles.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Sticky-boundary CK equations depend on the unproven delta-layer ansatz (4.9); a finite-width layer or a different epsilon-to-0 order of limits need not give (4.13)/(4.27).","rationale":"The reader's weakest assumption identifies the same load-bearing weakness: the sticky-boundary equations are derived by imposing the delta ansatz (4.9), which is a modeling postulate rather than a consequence of the underlying SDE. My independent reading confirms this, and it is the most consequential point in the paper because Section 4 is presented as a first-principles derivation of the encounter-based CK equations. I found the same issue in both the non-absorbing (4.13) and occupation-time (4.27) cases. I also noticed a likely sign error in (4.25c) that is inconsistent with the Laplace transform in (4.28c), and an apparently incomplete/incorrect equation (6.11) for the two-point correlation under global resetting; these are secondary but reinforce the conditional verdict. The finite-width-layer test would settle whether the boundary conditions are robust or ansatz-dependent. Since the reader already assigned CONDITIONAL, my read does not change the verdict.","tokens_in":25093,"tokens_out":19401,"duration_ms":173516,"concrete_test":"Implement a one-parameter family of boundary-layer regularizations compatible with the SDE (4.2): for layer width epsilon, let an RTP entering from 0+ be reflected (or mildly diffused) inside [-epsilon,0] until it tumbles at rate alpha and exits to x>0 with sigma=+1; compute the stationary or early-time boundary flux for epsilon = 1e-1, ..., 1e-4 by Monte Carlo and take epsilon to 0. Compare the limiting boundary condition with (4.13b)-(4.13c). Also repeat with the point-mass ansatz (4.9). If the two epsilon-to-0 limits differ, equations (4.13)/(4.27) rely on the specific ansatz and are not implied by the stochastic calculus. Additionally, correct (4.25c) to -q(0,t)delta(a) and confirm that (4.28c) follows; this checks the internal consistency of Section 4.2.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the generalized Ito lemma derives the sticky-boundary CK equations from sample-path dynamics is only as strong as the boundary-layer ansatz (4.9)/(4.23), which is assumed, not derived. In the SDE (4.2)/(4.17), a particle with X<0 has dX=0, so it remains at the point where it entered the layer (0-), not at x=-epsilon; the delta at x=-epsilon is an extra coordinate choice. A genuine layer of width epsilon would have a density q(t)/epsilon on [-epsilon,0], and the epsilon-to-0 limit of the averaged equations (or of a finite-epsilon boundary value problem) need not coincide with the limit obtained by inserting the delta ansatz before averaging. The derivation also contains an internal sign inconsistency: integration by parts in a gives the boundary term -q(0,t) on the right-hand side of (4.24), which requires -q(0,t)delta(a) in (4.25c); the printed plus sign is inconsistent with the Laplace transform (4.28c), which effectively uses the negative sign. Thus Sections 4.1 and 4.2 establish a conditional derivation: (4.13) and (4.27) follow only if (4.9) is accepted as the definition of the sticky boundary. The claimed rigorous mathematical framework is therefore overstated; the equations are a consequence of the ansatz plus the Ito calculus, not of the Ito calculus alone.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript develops a sample-path, stochastic-calculus framework for one-dimensional run-and-tumble particles, using a generalised Itô lemma for Poisson and diffusion processes to derive Chapman-Kolmogorov equations for standard motion, position/velocity resetting, and sticky boundaries. It then applies the framework to stochastic entropy production and to a population of globally resetting RTPs, and it derives exact non-equilibrium steady states in an appendix. The appendix recovers the Evans-Majumdar result in the symmetric-resetting case, and the sticky-boundary equations reproduce the author's earlier equations from Ref. [15].","tokens_in":25477,"tokens_out":12190,"duration_ms":106688,"significance":"If the derivation is made fully rigorous, the paper offers a useful unified route from RTP sample paths to CK equations, including encounter-based sticky boundaries, and it introduces a population-level SPDE whose global-resetting correlations and moment hierarchy are genuinely interesting. The stochastic-entropy analysis and the explicit NESS formulas are also valuable. However, the central sticky-boundary derivation depends on an assumed delta-layer ansatz rather than on the stated stochastic calculus alone, and several equations contain sign or factor errors. The paper's main contribution is therefore currently a promising program with important local corrections needed.","major_comments":[{"comment":"The boundary-layer ansatz rho_{epsilon,k}(x,t)=rho_k(x,t)1_{x>0}+delta(x+epsilon)delta_{k,-1}q(t) is assumed, not derived from the SDE (4.2). In that SDE a particle with X<0 has dX=0, so it remains at the point where it entered the layer rather than at x=-epsilon; the delta at x=-epsilon is an extra coordinate choice. If a finite-width layer has density q(t)/epsilon on [-epsilon,0], the epsilon-to-0 limit of the averaged equations need not equal the limit obtained by inserting the delta ansatz before averaging. Since Eqs. (4.13) and (4.27) are the main new results of Section 4, the paper must either state (4.9) explicitly as a modeling definition of the sticky state and soften the claim of a rigorous derivation, or provide a genuine limiting argument from a finite-width layer.","section":"Section 4.1, Eq. (4.9)"},{"comment":"The printed plus sign in front of q(0,t)delta(a) in Eq. (4.25c) is inconsistent with the integration-by-parts calculation in Eq. (4.24), which gives a boundary term -f(0-,0,-1)q(0,t) and hence should yield -q(0,t)delta(a) in (4.25c). The Laplace-transformed equation (4.28c) is also consistent with the negative sign, not the printed plus sign. The sign must be corrected, or the discrepancy explained; as written, the SPDE (4.25c) does not reproduce the boundary value problem (4.28) that is subsequently used.","section":"Section 4.2, Eq. (4.25c) vs. Eq. (4.28c)"},{"comment":"The coefficient of the white-noise term in Eq. (6.5) is -2 sqrt(D)/M sum_j xi_j(t) partial_x delta(x-X_j(t)), but the SDE (6.1) has sqrt(2D) dW_j(t), so the correct coefficient is -sqrt(2D)/M sum_j ... = -sqrt(2) sqrt(D)/M sum_j ... . This is a factor-of-sqrt(2) error in a stated SPDE. Although the term has zero mean and therefore does not affect the averaged equation (6.8), it would affect any fluctuation-level statement derived from (6.5), so it must be fixed.","section":"Section 6, Eq. (6.5)"},{"comment":"The sign of the resetting term in Eq. (6.16) is wrong. From Eq. (6.15b), subtracting the equations for k=1 and k=-1 gives dZ0/dt=-2alpha Z0 + h(t)[lambda_-^(0)-Z0(t-)], not -h(t)[lambda_-^(0)-Z0(t-)]. As printed, a reset event maps Z0 to 2Z0-lambda_-^(0) instead of to lambda_-^(0). The subsequent solution (6.17)-(6.19) uses the correct positive sign, so the inconsistency is localized but should be corrected.","section":"Section 6, Eq. (6.16)"},{"comment":"Eq. (5.58) states Rsys(t)-Rsys(t)=alpha integral ... >=0, which is identically zero on the left and cannot be positive on the right. This is not a meaningful second-law statement as printed. The intended left-hand side presumably involves a different quantity such as Rtot(t)-Rsys(t) or Rsys(t)-Rres(t); the authors should provide the correct equation and define all quantities appearing in it.","section":"Section 5.2, Eq. (5.58)"}],"minor_comments":[{"comment":"The Laplace-transform definition in Eq. (4.29) contains notation errors: the second line writes eQ(z,s) on the left but averages over a on the right using eQ(a,t), and it should define eQ(z,t)=int_0^infty e^{-za}Q(a,t)da. Please correct the notation.","section":"Section 4.2, Eq. (4.29)"},{"comment":"There is a missing closing parenthesis in the denominator terms p_{sigma(t)}(X(t,t); these should read p_{sigma(t)}(X(t),t).","section":"Section 5.1, Eq. (5.39)"},{"comment":"The right-hand side of Eq. (6.15b) contains the typo M 0)_k(t); this should be M_k^{(0)}(t).","section":"Section 6, Eq. (6.15b)"},{"comment":"The text after Eq. (2.10) says 'in terms of the Poison process'; this should be 'Poisson process'.","section":"Section 2, text near Eq. (2.10)"},{"comment":"The sentence 'Performing the various steps outlined at the end of section 2, we obtain...' does not show the derivation; given that the resetting terms interact with the sticky boundary and the bound state, it would be clearer to display at least the SPDE or the key averaging step that produces Eqs. (4.32a)-(4.32c).","section":"Section 4.3"}],"recommendation":"major_revision","confidential_remarks":"The paper overlaps substantially with the author's previous work in Refs. [14,15], and the genuinely new material is the unified stochastic-calculus derivation plus the population-level global-resetting analysis. The errors listed above are technical and fixable, so I do not recommend rejection, but the sticky-boundary derivation needs to be reframed as a modelling assumption or supplied with a rigorous limit before the paper's central claim is accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a solid consolidation of Bressloff's existing program, not a paradigm shift. The genuinely new pieces are the boundary-layer derivation of the occupation-time propagator (4.27), the population SPDE with global resetting (6.8), and the moment hierarchy with embedded lower-order resetting dynamics. Those are worth having, and they are derived with real care.\n\nThe generalized Ito lemma is standard, and large parts of Sections 2 and 3 rederive known CK equations. That is fine as review material, but it is not the contribution. The entropy section also recovers known results from Paoluzzi et al. and Evans-Majumdar, including the NESS in the appendix; the checks against external results are honest and the algebra appears correct.\n\nThe soft spots are real but not fatal. Most important: the sticky-boundary equations (4.13) and (4.27) follow only if you accept the delta-layer ansatz (4.9)/(4.23), where the bound particle sits at a single point x = -epsilon. That is assumed, not derived. A finite-width layer, or a different order of limits, need not give the same boundary conditions. The paper's claim to provide a rigorous derivation is therefore overstated; it provides a conditional derivation from the ansatz plus Ito calculus. Relatedly, there is a sign inconsistency in (4.25c): integration by parts gives -q(0,t)delta(a), and the Laplace transform (4.28c) effectively uses the negative sign, but the printed equation has a plus. This needs correcting.\n\nElsewhere, Eq. (6.5) has a factor of 2 in front of the sqrt(D) noise term that does not match the SDE (6.1); it drops out after averaging, so it is minor. Eq. (5.58) repeats Rsys(t) on both sides, which is a typo, and Section 4.3 explicitly defers the resetting-with-occupation-time derivation, so that part is unfinished rather than wrong. None of these undermine the central message, but they do mean the manuscript is not ready as-is.\n\nWho is this for? Someone working on run-and-tumble particles with resetting or sticky boundaries will get real value from the consolidated framework and from the new population-level correlation result. I would send it to peer review, with a request to fix the sign error, clarify or relax the boundary-layer ansatz, and correct the minor typos. A serious referee can handle that.","headline":"A useful consolidation of the author's own RTP program: the sticky-boundary and global-resetting sections are the real content, but the sticky-boundary derivation rests on an unproven delta-layer ansatz and contains a sign inconsistency that should be fixed.","tokens_in":25937,"tokens_out":1458,"would_cite":true,"duration_ms":14622,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["60H10","60J76","82C31"],"pacs":["05.40.-a","05.60.-k"],"model":"deepseek-v4-flash","headline":"The paper claims that a generalized Itô lemma for jump and diffusion processes gives a direct sample-path route to the Chapman-Kolmogorov equations of one-dimensional run-and-tumble motion, including resetting, sticky boundaries, entropy…","keywords":["run-and-tumble particle","stochastic calculus","Itô lemma","Chapman-Kolmogorov equation","stochastic resetting","sticky boundary","stochastic entropy production","global resetting"],"falsifier":"Numerically integrate the boundary-layer SDE for a sticky wall with a finite layer width $\\epsilon$ and a spatially distributed bound state, and test whether the flux relation $\\alpha Q(t)=v p_1(0^+,t)$ together with $dQ/dt=-\\alpha Q+v p_{-1}(0^+,t)$ holds as $\\epsilon\\to 0$; a persistent discrepancy would show the delta-function boundary-layer ansatz is not a consequence of the stochastic calculus.","tokens_in":24913,"feed_emoji":"🏃","tokens_out":10767,"duration_ms":92111,"temperature":0.7,"pith_summary":"Run-and-tumble motion is usually studied through its Chapman-Kolmogorov equation for the joint probability of position and velocity state. This paper develops the opposite route: model the velocity switches and any resetting events as Poisson processes, write the position as a jump-diffusion, and apply a generalized Itô lemma to the empirical measure $\\rho_k(x,t)=\\delta(x-X(t))\\delta_{k,\\sigma(t)}$. The resulting stochastic partial differential equation, when averaged over the noise sources, is exactly the CK equation. The author uses this pipeline to derive resetting and sticky-boundary equations, to define stochastic entropy along individual trajectories, and to show that global resetting makes noninteracting particles statistically correlated.","feed_headline":"One Itô-type lemma links run-and-tumble paths to master equations","feed_subtitle":"One averaging step turns sample paths into the master equations for resetting, sticky walls, and populations.","key_machinery":"The workhorse is the generalised Itô lemma for jump-diffusions driven by Brownian motion and Poisson processes: for a test function $f$, $df = (v\\sigma f' + D f'')dt + \\sqrt{2D}f'dW + [f(X,-\\sigma)-f(X,\\sigma)]dN$, plus resetting jump terms when present. Its role is to convert sample-path dynamics into an SPDE for the empirical measure; averaging that SPDE over the independent noise sources, using the Poisson-process independence identity $\\mathbb{E}[F(X(t^-),\\sigma(t^-))dN(t)]=\\alpha\\,dt\\,\\mathbb{E}[F]$, is the single step that produces every Chapman-Kolmogorov equation in the paper.","core_discovery":"The central claim is that for a one-dimensional run-and-tumble particle with dynamics $dX=v\\sigma\\,dt+\\sqrt{2D}\\,dW$ plus jump terms and $d\\sigma=-2\\sigma(t^-)dN(t)$, with an independent resetting Poisson process when present, the generalised Itô lemma determines everything. Applied to the empirical measure, it yields an SPDE whose expectation reproduces the forward CK equation $\\partial_t p_k = -v k\\partial_x p_k + D\\partial_x^2 p_k + \\alpha(p_{-k}-p_k)$ together with the resetting and boundary terms. The same machinery re-derives, rather than postulates, the occupation-time propagator equations for a partially absorbing sticky boundary and the stochastic CK equation for a population with global resetting. The paper also establishes that the pathwise stochastic entropy defined by $S_{\\rm sys}(t)=-\\ln p_{\\sigma(t)}(X(t),t)$ averages to the Gibbs-Shannon entropy, with steady-state total entropy production $v^2/D$ in the purely diffusive case.","pith_inferences":["Editorial inference: the same SPDE-averaging pipeline should extend to biased runs or position-dependent switching rates, because the Poisson-jump calculus does not use symmetry between the two velocity states.","Editorial inference: the boundary-layer ansatz suggests a concrete modelling test—comparing the derived flux relation $\\alpha Q(t)=v p_1(0^+,t)$ against simulations with a finite-width sticky zone would show how much resolution the point-localization limit actually retains.","Editorial inference: the global-resetting correlation mechanism could be observed in experiments where active colloids or bacteria are synchronously returned to starting conditions by an external global pulse; the predicted two-particle covariance is a measurable signature that does not require interactions.","Editorial inference: the embedded moment hierarchy hints at a closure problem for population statistics under global resetting; a possible extension is to ask whether a truncated set of moments with resetting reproduces the full density statistics in some limit."],"forward_implications":["For an RTP with diffusion and resetting, averaging the empirical-measure SPDE gives the CK equation (3.13), recovering the standard resetting master equation when $D=0$.","The sticky-boundary CK equations, including the occupation-time propagator for absorption at a threshold, follow from the boundary-layer ansatz rather than being assumed; the Laplace-transformed equations reproduce the encounter-based model introduced earlier heuristically.","Along individual trajectories the stochastic entropy production rate averages to the Gibbs-Shannon entropy rate $\\frac{d}{dt}S_{\\rm GS}(t)$, so the second law appears only after ensemble averaging.","For a population of noninteracting RTPs, a global resetting clock makes $\\mathbb{E}[\\Phi_j(x,t)\\Phi_k(y,t)]\\ne \\mathbb{E}[\\Phi_j(x,t)]\\,\\mathbb{E}[\\Phi_k(y,t)]$, so global resetting alone creates particle correlations.","The moment equations of the population density form a hierarchy of ODEs with resetting in which lower-order moments are embedded in higher-order equations, so joint distributions of low-order moments are needed at each level."],"supporting_citations":[{"why":"Supplies the sticky-boundary occupation-time propagator equations that the stochastic calculus re-derives from first principles, and the finite-interval solution used for comparison.","marker":"[15]"},{"why":"Provides the renewal-method non-equilibrium stationary state for an RTP under resetting, which the appendix extends to asymmetric velocity resetting and uses as a check.","marker":"[17]"},{"why":"Introduces the encounter-based model of diffusion-mediated absorption, including Brownian functionals and random occupation-time thresholds, that motivates the sticky-boundary absorption model.","marker":"[20, 21, 22, 23]"},{"why":"Gives the Gibbs-Shannon entropy rate and steady-state $v^2/D$ entropy production with which the pathwise stochastic-entropy averages are matched.","marker":"[33]"},{"why":"Provides the global density equation framework for particle systems with stochastic resetting that the population-level SPDE generalizes.","marker":"[36]"},{"why":"Reviews global resetting in interacting particle systems and supplies the shared reset-clock protocol used for the population model.","marker":"[38]"}],"fun_headline_variants":["Itô's lemma unifies run-and-tumble, resetting, and sticky walls","A single Itô rule yields run-and-tumble master equations","From paths to equations: one Itô lemma for RTP dynamics","Itô lemma unifies run-and-tumble, resetting, and boundaries"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that a particle stuck at a sticky wall can be treated as a point mass localised at a single boundary-layer point, with the layer width sent to zero before averaging over the Poisson switching process; if the bound state has finite spatial extent or position-dependent tumbling, the derived sticky-boundary equations are not implied.","fun_headline_variants_meta":{"raw":{"variants":["Itô's lemma unifies run-and-tumble, resetting, and sticky walls","A single Itô rule yields run-and-tumble master equations","From paths to equations: one Itô lemma for RTP dynamics","Itô lemma unifies run-and-tumble, resetting, and boundaries"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00116,"raw_usage":{"total_tokens":4834,"prompt_tokens":1005,"completion_tokens":3829,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":621,"completion_tokens_details":{"reasoning_tokens":3749}},"tokens_in":621,"tokens_out":3829,"duration_ms":25818,"temperature":1.0,"reasoning_tokens":3749,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:10:54.262928+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Numerically integrate the boundary-layer SDE for a sticky wall with a finite layer width $\\epsilon$ and a spatially distributed bound state, and test whether the flux relation $\\alpha Q(t)=v p_1(0^+,t)$ together with $dQ/dt=-\\alpha Q+v p_{-1}(0^+,t)$ holds as $\\epsilon\\to 0$; a persistent discrepancy would show the delta-function boundary-layer ansatz is not a consequence of the stochastic calculus.","supporting_citations":[{"cited_title":"2023 Encounter-based model of a run-and-tumble particle II: ab- sorption at sticky boundaries","cited_arxiv_id":null,"evidence_quote":"Supplies the sticky-boundary occupation-time propagator equations that the stochastic calculus re-derives from first principles, and the finite-interval solution used for comparison."},{"cited_title":"2018 Run and tumble particle under resetting: a renewal approach","cited_arxiv_id":null,"evidence_quote":"Provides the renewal-method non-equilibrium stationary state for an RTP under resetting, which the appendix extends to asymmetric velocity resetting and uses as a check."},{"cited_title":"2024 Entropy production of run-and-tumble particles","cited_arxiv_id":null,"evidence_quote":"Gives the Gibbs-Shannon entropy rate and steady-state $v^2/D$ entropy production with which the pathwise stochastic-entropy averages are matched."},{"cited_title":"2024 Global density equations for interacting particle systems with stochastic resetting: from overdamped Brownian motion to phase synchronization Chaos 34, 043101","cited_arxiv_id":null,"evidence_quote":"Provides the global density equation framework for particle systems with stochastic resetting that the population-level SPDE generalizes."},{"cited_title":"2023 Stochastic resetting in interacting particle systems: a review’ J","cited_arxiv_id":null,"evidence_quote":"Reviews global resetting in interacting particle systems and supplies the shared reset-clock protocol used for the population model."}],"review_version":1}