{"id":"a8770120-552e-4e68-bcd5-bd3ef981cb18","arxiv_id":"2606.20234","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":4.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"A conservative adaptive rank method for the Wigner-Poisson system that preserves mass, momentum, and energy invariants via local macroscopic correction plus global quadratic moment adjustment.","lead":"The paper proposes a conservative adaptive rank method for the 1D1V Wigner-Poisson system that pairs adaptive rank compression with local density-momentum updates and a global energy correction using Fermi-Dirac reconstruction. A smart generalist might read it to see how numerical methods can cut the cost of quantum plasma simulations while keeping key physical quantities accurate.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"Fermi-Dirac reconstruction plus global energy correction may leave residual phase-space errors after ACA SVD compression that affect long-time dynamics","rationale":"The reader's weakest assumption correctly isolates the reconstruction-correction interface under active rank adaptation as the least-secured link in the conservation argument. The numerical evidence cited is post-hoc and does not isolate whether truncation re-opens the invariants; the proposed check directly measures that gap without requiring new physics.","tokens_in":1828,"tokens_out":391,"duration_ms":12955,"concrete_test":"At each time step of the two-stream instability run, store the kinetic solution immediately after the global quadratic moment correction but before ACA SVD truncation; recompute the discrete mass, momentum, and energy from this pre-compression state and again after truncation to the adaptive rank chosen by the ACA tolerance. If the post-truncation energy deviates from the target by more than 10× machine epsilon while the pre-truncation value matches to machine precision, the compression step is the source of the residual error.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on the local density-momentum update being transferred via Fermi-Dirac reconstruction and then globally corrected for total energy, after which the result is folded into an ACA SVD representation whose rank adapts. Nothing in the described procedure guarantees that the post-correction distribution remains consistent with the macroscopic moments once the low-rank factors are truncated; any truncation error re-introduces discrepancies that the global quadratic correction (applied only once per step) cannot retroactively cancel. Because the Wigner operator is nonlocal, even small moment violations can accumulate in the self-consistent field, especially for varying H. The reported near-machine-precision conservation is an empirical outcome on three specific 1D1V periodic tests; it does not demonstrate that the reconstruction-correction-compression sequence is algebraically closed with respect to the discrete invariants.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript proposes a conservative adaptive rank method for the 1D1V Wigner-Poisson system. It combines a sampling-based adaptive rank Wigner-Poisson update with a macroscopic correction consisting of a conservative density-momentum solve, Fermi-Dirac-type reconstruction to transfer local updates to the kinetic level, and a global quadratic moment correction to enforce the discrete total energy constraint. The corrected distribution is folded into an ACA SVD representation whose rank adapts to the complexity induced by the nonlocal Wigner operator and self-consistent Poisson field. Numerical experiments on the two-stream instability, strong Landau damping, and bump-on-tail instability for several values of the quantum parameter H demonstrate that the method captures benchmark dynamics, maintains bounded adaptive ranks, and preserves the specified global invariants with errors near machine precision. A comparison with a related globally conservative formulation is also presented, showing nearly identical results in the tested periodic setting.","tokens_in":1998,"tokens_out":564,"duration_ms":17991,"significance":"If the reported conservation properties are robust under adaptive-rank compression, the approach would provide a practical route to long-time deterministic simulations of quantum kinetic systems at reduced cost while retaining the macroscopic invariants required for physical fidelity. The explicit comparison of local-plus-global versus fully global correction strategies and the observation of bounded ranks across multiple instabilities constitute concrete strengths.","major_comments":[{"comment":"Numerical experiments section: the abstract and results claim conservation errors reach near machine precision and that benchmark dynamics are captured, yet no quantitative error tables, convergence studies with respect to rank tolerance or time step, or plots of invariant drift versus H are supplied; without these data the central claim that the reconstruction-correction-compression sequence reliably preserves the discrete invariants cannot be assessed.","section":"Numerical experiments"},{"comment":"Method section (description of the correction and ACA SVD step): the Fermi-Dirac reconstruction followed by a single global quadratic energy correction is applied before the ACA SVD truncation; no algebraic argument or numerical test demonstrates that the post-truncation distribution remains consistent with the corrected macroscopic moments, leaving open the possibility that truncation re-introduces moment errors that subsequently affect the self-consistent field, especially for varying H.","section":"Method section"}],"minor_comments":[{"comment":"The notation for the adaptive rank tolerance and the precise definition of the quadratic moment correction operator should be stated explicitly rather than referenced only to prior work.","section":null},{"comment":"Figure captions for the phase-space plots should include the specific H values and final simulation times to allow direct comparison with the conservation-error statements.","section":null}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the careful reading and constructive comments. We address the two major comments point by point below, indicating the revisions we will make.","responses":[{"response":"We agree that the presentation would benefit from additional quantitative data. In the revised manuscript we will insert tables reporting the maximum conservation errors for mass, momentum, and energy over the full simulation interval for each benchmark and each tested value of H. We will also add plots of the time histories of the invariant drifts versus H and a short convergence study with respect to the ACA rank tolerance (showing that errors remain near machine precision down to the tolerances used in the main experiments). These additions will directly support the central conservation claim.","revision_made":"yes","referee_comment":"Numerical experiments section: the abstract and results claim conservation errors reach near machine precision and that benchmark dynamics are captured, yet no quantitative error tables, convergence studies with respect to rank tolerance or time step, or plots of invariant drift versus H are supplied; without these data the central claim that the reconstruction-correction-compression sequence reliably preserves the discrete invariants cannot be assessed."},{"response":"The reported numerical results already show that the complete sequence (correction followed by ACA SVD) yields invariant errors at or near machine precision for all tested H. This empirical evidence indicates that any re-introduced moment errors remain negligible under the chosen rank tolerances. Nevertheless, we acknowledge the absence of an a-priori algebraic guarantee. In the revision we will add a short paragraph explaining that the ACA tolerance is set below the target conservation threshold and will include a supplementary numerical check (for one representative case) that compares the macroscopic moments immediately before and after the SVD truncation step. A rigorous algebraic proof for the adaptive-rank case lies outside the present scope.","revision_made":"partial","referee_comment":"Method section (description of the correction and ACA SVD step): the Fermi-Dirac reconstruction followed by a single global quadratic energy correction is applied before the ACA SVD truncation; no algebraic argument or numerical test demonstrates that the post-truncation distribution remains consistent with the corrected macroscopic moments, leaving open the possibility that truncation re-introduces moment errors that subsequently affect the self-consistent field, especially for varying H."}],"tokens_in":1544,"tokens_out":484,"duration_ms":19950,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The core idea is straightforward: run the sampling-based adaptive rank update from earlier work, then apply a local density-momentum correction, reconstruct with a Fermi-Dirac form, enforce total energy globally, and compress again with ACA SVD. The authors show that this keeps the adaptive rank bounded and produces results close to a fully global conservative variant on the two-stream, strong Landau damping, and bump-on-tail problems for several values of H.\n\nWhat stands out is the explicit comparison between the local-plus-global correction and the earlier global-only approach; both give nearly identical phase-space and diagnostic outputs in the periodic 1D1V setting. That is useful information for anyone already using low-rank methods on this system.\n\nThe main limitation is that the abstract gives no tables, no convergence rates, and no breakdown of how truncation error after the correction step affects the moments. The stress-test concern about residual phase-space discrepancies reappearing after ACA compression is therefore still open; the Wigner operator is nonlocal, so even small moment drift could matter over long times. If the full paper contains the missing quantitative checks and they hold, the method is on firmer ground.\n\nThis is aimed at people already working on deterministic quantum kinetic codes who need cheaper phase-space representations while keeping invariants. A reader in that niche will get concrete implementation details and a direct head-to-head with prior correction strategies. The work is concrete enough and the tests are standard enough that it deserves a serious referee.","headline":"The paper adds Fermi-Dirac reconstruction plus a global energy correction to an existing adaptive-rank scheme for 1D1V Wigner-Poisson and reports near-machine-precision conservation on three standard tests.","tokens_in":2552,"tokens_out":383,"would_cite":false,"duration_ms":16116,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"A conservative adaptive rank method for the Wigner-Poisson system combines local density-momentum updates with global energy correction to preserve invariants while adapting numerical rank.","keywords":["Wigner-Poisson system","adaptive rank","conservative correction","Fermi-Dirac reconstruction","numerical methods for kinetic equations","quantum parameter H","instability benchmarks"],"falsifier":"A run of one of the benchmark instabilities in which the discrete total energy deviates from machine precision by more than a few orders of magnitude while the adaptive rank procedure is engaged.","tokens_in":2748,"feed_emoji":"","tokens_out":698,"duration_ms":23244,"temperature":0.7,"pith_summary":"The paper develops a numerical scheme for the one-dimensional one-velocity Wigner-Poisson system that lowers the cost of phase-space evolution by letting the representation rank change with local complexity. It pairs a sampling-based adaptive rank update with a macroscopic correction step that first solves for density and momentum locally, transfers those values through a Fermi-Dirac reconstruction, and then applies a global quadratic moment correction to enforce total energy. Experiments on the two-stream instability, strong Landau damping, and bump-on-tail instability demonstrate that the resulting scheme reproduces known benchmark behavior for several values of the quantum parameter while keeping the adaptive ranks bounded and holding the discrete invariants to near machine precision.","feed_headline":"Adaptive rank method holds Wigner-Poisson invariants to machine precision","feed_subtitle":"Local density-momentum updates plus global energy correction keep conservation errors near machine precision while ranks stay bounded in sta","key_machinery":"The conservative macroscopic correction that performs local density-momentum updates through Fermi-Dirac reconstruction followed by a global quadratic moment correction for total energy.","core_discovery":"The scheme merges a sampling-based adaptive rank Wigner-Poisson update with a conservative macroscopic correction that supplies local density-momentum updates, uses a Fermi-Dirac-type reconstruction to map them onto the kinetic solution, and adds a global quadratic moment correction to satisfy the discrete total energy constraint. The corrected state is then stored in an ACA SVD representation whose rank adapts to the phase-space structure produced by the nonlocal Wigner operator and the self-consistent Poisson field.","pith_inferences":["The near-equivalence of local-plus-global and fully global correction strategies suggests that designers of adaptive-rank schemes have latitude in choosing where to enforce conservation.","Because the reconstruction is motivated by the quantum-statistical structure of the Wigner-Poisson model, similar corrections may prove useful in other quantum kinetic equations whose equilibrium distributions differ from classical Maxwell-Boltzmann forms."],"forward_implications":["The method reproduces the expected phase-space evolution in the two-stream instability, strong Landau damping, and bump-on-tail instability for multiple values of the quantum parameter.","Adaptive ranks remain bounded throughout the evolution of the tested periodic problems.","Global discrete invariants are preserved with errors near machine precision.","Results are nearly identical to those obtained from a related formulation that enforces mass, momentum, and energy globally."],"fun_headline_variants":["Adaptive rank conserves Wigner-Poisson invariants","Conservative correction adapts Wigner-Poisson ranks","Global energy correction enables adaptive Wigner-Poisson","Rank adapts while conserving Wigner-Poisson moments"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The Fermi-Dirac reconstruction transfers local macroscopic updates to the kinetic level without leaving errors that the subsequent global energy correction cannot remove, even when adaptive rank compression is active.","fun_headline_variants_meta":{"raw":{"variants":["Adaptive rank conserves Wigner-Poisson invariants","Conservative correction adapts Wigner-Poisson ranks","Global energy correction enables adaptive Wigner-Poisson","Rank adapts while conserving Wigner-Poisson moments"]},"model":"grok-4.3","cost_usd":0.007441,"raw_usage":{"total_tokens":3466,"prompt_tokens":764,"num_sources_used":0,"completion_tokens":61,"cost_in_usd_ticks":74412000,"prompt_tokens_details":{"text_tokens":764,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":2641,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":764,"tokens_out":61,"duration_ms":20143,"temperature":1.0,"reasoning_tokens":2641,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-26T16:21:55.551811+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"A run of one of the benchmark instabilities in which the discrete total energy deviates from machine precision by more than a few orders of magnitude while the adaptive rank procedure is engaged.","supporting_citations":[],"review_version":1}