{"id":"2b44f759-caf1-4dff-b675-c1e94342082b","arxiv_id":"2607.12613","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":5.5,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Tensor-train reduced-order models of the parameter-to-observation map enable cheaper, more robust optimization-based inversion for parameter-dependent dynamical systems.","lead":"A tensor-train reduced-order model approximates the parameter-to-observation map for inverse problems so optimization can run in reduced coordinates at lower online cost. It may matter for engineers and scientists who need faster, more robust parameter recovery from noisy dynamical-system data.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified beyond the abstract-only information limit already flagged by the Reader.","rationale":"The Reader’s UNVERDICTED / LOW-confidence stance is the only defensible position given an abstract-only review of a numerical-methods paper. The strongest claim is computational and empirical; its soundness rests on the TT compressibility assumption the Reader already named. No further load-bearing flaw (circularity, incorrect derivation, mismatched baselines) can be diagnosed without the missing equations, tables, or code. Therefore the stress-test finds no reason to alter the verdict, the confidence, or the identified weakest assumption. The concrete test above simply operationalizes the check that would be performed once the full manuscript is available.","tokens_in":2100,"tokens_out":398,"duration_ms":3510,"concrete_test":"When the full PDF appears, extract the TT ranks and relative approximation errors reported for the parameter-to-observation map in both example problems (especially the highest-dimensional inclusion case). Recompute or re-plot the inverse-solution error versus those ranks; if the inverse error remains comparable to full-order inversion only for ranks that grow faster than the claimed online-cost savings, the compressibility assumption fails to support the central claim.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The Reader correctly isolates TT compressibility of the parameter-to-observation map as the load-bearing assumption for both accuracy and the reduced-coordinate Gauss–Newton reformulation. With only the abstract available, no equations, rank tables, error-vs-rank curves, or Jacobian-assembly timings can be inspected, so no additional internal inconsistency or hidden assumption can be verified or refuted. The experimental program described (heat-transfer inclusions + FitzHugh–Nagumo, noise/regularization/dimension sweeps) is the appropriate place to test that assumption; its absence simply leaves the claim uncheckable rather than contradicted.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript develops a tensor reduced-order modeling (TROM) framework for optimization-based inverse problems governed by parameter-dependent dynamical systems. The parameter-to-observation map is approximated in tensor-train format via TT-SVD or TT-Cross and embedded in a regularized nonlinear least-squares formulation. The low-rank structure is further used to reformulate the inverse problem in reduced coordinates, assemble Gauss–Newton quantities without forming the full observation-space Jacobian, and perform TROM-based objective minimization over a discrete parameter grid (stand-alone or as initialization for Gauss–Newton). Two applications are claimed: inverse heat transfer with multiple low-conductivity inclusions, and FitzHugh–Nagumo parameter estimation with a highly nonconvex landscape. Numerical experiments are said to assess ROM error, noise, regularization, initialization, discretization, and parameter dimension, concluding that TROM reproduces full-order inversion behavior at substantially reduced online cost and that reduced-coordinate inversion, tensor optimization, and regularization improve robustness in higher-dimensional, noisy, and nonconvex regimes.","tokens_in":2151,"tokens_out":1062,"duration_ms":21085,"significance":"If substantiated, the contribution is a coherent computational methodology that couples TT compression of the parameter-to-observation map with reduced-coordinate Gauss–Newton and discrete tensor optimization for inverse problems. The dual use of the tensor representation—for both forward acceleration and inverse reformulation—is a clear methodological strength relative to pure surrogate-forward ROMs. The stated experimental program (heterogeneous heat transfer and FitzHugh–Nagumo; sweeps over ROM error, noise, regularization, initialization, discretization, and dimension) is well matched to the claims and would be of genuine interest to the inverse-problems and reduced-order modeling communities if the online-cost and robustness claims hold with documented evidence.","major_comments":[{"comment":"The central accuracy claim depends on the parameter-to-observation map admitting a sufficiently accurate low-rank TT representation so that ROM error does not destroy the inverse solution. The abstract asserts that ROM approximation error is assessed, but without quantitative rank/tolerance data, error-vs-rank (or error-vs-tolerance) curves, and inverse-error-vs-compression results, this load-bearing assumption cannot be verified. The manuscript must report the TT ranks and compression tolerances used and show that inverse reconstructions remain faithful as compression is varied.","section":"Abstract (TT compressibility / ROM error)"},{"comment":"The claim that Gauss–Newton quantities can be assembled without forming the full observation-space Jacobian is load-bearing for the reduced-coordinate inversion story and for the asserted online-cost reduction. The abstract does not state the algorithmic construction, complexity, storage, or accuracy of this assembly relative to full-order Jacobian formation. The full manuscript must supply the derivation and numerical timings that show a genuine online saving rather than a shift of cost into offline TT construction.","section":"Abstract (Gauss–Newton reformulation)"},{"comment":"The headline empirical claims—that TROM reproduces full-order inversion behavior at substantially reduced online cost, and that reduced-coordinate inversion, tensor optimization, and regularization improve robustness under noise, higher dimension, and nonconvexity—require side-by-side metrics (reconstruction error, residual, iteration counts, wall-clock online cost) against full-order inversion across the stated sweeps. Without those tables and figures the claims remain uncheckable from the abstract alone.","section":"Abstract (experimental claims)"}],"minor_comments":[{"comment":"Both TT-SVD and TT-Cross are listed, but the abstract does not indicate which is used for which experiment or how the choice affects inverse accuracy; a brief early clarification would help.","section":"Abstract"},{"comment":"The phrase “TROM-based objective minimization over the discrete parameter grid” should be defined more precisely early (e.g., exhaustive search on a TT-compressed objective, ALS-type TT optimization, or another scheme).","section":"Abstract"},{"comment":"“Substantially reduced online cost” should be accompanied by a clear offline/online cost split so that readers can judge when the method is advantageous.","section":"Abstract"}],"recommendation":"uncertain","confidential_remarks":"Only the abstract was available for this review; the full text was not provided. I therefore cannot inspect equations, rank tables, error curves, Jacobian timings, or reconstruction figures. Recommendation is uncertain pending the full manuscript. The Reader correctly identifies TT compressibility of the parameter-to-observation map as the load-bearing assumption; the Skeptic finds no further internal inconsistency beyond that abstract-only limit, and I agree. If the full paper supplies the quantitative evidence outlined in the major comments, the work appears methodologically coherent and likely suitable for minor or major revision rather than rejection. Scope and experimental design look appropriate for a computational inverse-problems / numerical-analysis venue."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"Punchline: this is a methods paper that wires tensor-train surrogates into an optimization-based inverse solver for parametric dynamical systems. The claim is that you get essentially full-order inversion behavior at substantially lower online cost, plus better robustness in noisy and nonconvex regimes. We only have the abstract, so that claim is plausible but not yet inspectable.\n\nWhat is actually new is the integrated pipeline rather than any single ingredient. They approximate the parameter-to-observation map directly in TT format (TT-SVD or TT-Cross), reformulate the inverse problem in the reduced coordinates so the Gauss–Newton quantities can be assembled without ever forming the full observation-space Jacobian, and add a tensor-based objective minimization over a discrete parameter grid that can either stand alone or warm-start the subsequent solve. The two test problems—heterogeneous heat transfer with multiple low-conductivity inclusions, and FitzHugh–Nagumo with a highly nonconvex landscape—are sensible choices, and the experimental program they describe (ROM error, noise, regularization, initialization, discretization, parameter dimension) is exactly the right checklist.\n\nIf the full paper delivers the error-vs-rank curves, timing tables, and fair full-order baselines that the abstract promises, this is a useful computational contribution for people already working with TT/ROM and inverse problems. No structural circularity is visible; the free parameters (TT ranks/tolerance, regularization, grid resolution) are the usual ones for the field.\n\nThe soft spot is load-bearing and correctly flagged by the reader: everything rests on the parameter-to-observation map admitting a sufficiently accurate low-rank TT representation. If that fails, both accuracy and the reduced-coordinate Jacobian assembly break. With only the abstract we cannot see ranks, approximation errors, or whether the reduced solves actually recover the full-order solutions. That is an information limit, not an internal contradiction.\n\nThis is for numerical analysts and inverse-problems practitioners who already use or are open to tensor trains. It is not foundational theory. It deserves a serious referee rather than a desk reject; if the PDF has the promised experiments and enough detail to reproduce the ranks and timings, send it out.","headline":"Plausible TROM pipeline for inverse problems that claims full-order accuracy at lower online cost; abstract-only, so the numerics and TT ranks are uncheckable.","tokens_in":2827,"tokens_out":541,"would_cite":false,"duration_ms":13542,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["65M32","15A69","65F55","65K10","49N45"],"pacs":[],"model":"grok-4.5","headline":"A tensor-train reduced-order model of the parameter-to-observation map recovers full-order inverse solutions for dynamical systems at substantially lower online cost.","keywords":["tensor-train","reduced-order modeling","inverse problems","Gauss-Newton","parameter estimation","dynamical systems","TT-SVD","TT-Cross"],"falsifier":"Identify or construct a dynamical system whose parameter-to-observation map has high tensor-train rank; run TROM and full-order inversion on identical noisy data and check whether TROM recovers parameters that match the full-order solution within the paper’s reported error tolerances—systematic mismatch falsifies the claim that TROM reproduces full-order behavior.","tokens_in":2910,"feed_emoji":"⚡","tokens_out":1035,"duration_ms":20448,"temperature":0.7,"pith_summary":"This paper builds a tensor reduced-order modeling (TROM) framework that approximates the parameter-to-observation map of a dynamical system directly in tensor-train format, using TT-SVD or TT-Cross compression, and folds that representation into a regularized nonlinear least-squares inverse problem. The low-rank structure is used not only to speed forward evaluations but also to rewrite the inversion in reduced coordinates, assemble Gauss–Newton quantities without the full observation-space Jacobian, and minimize the TROM objective over a discrete parameter grid—either as a stand-alone solve or as a data-informed warm start. On an inverse heat-transfer problem with multiple low-conductivity inclusions and on a highly nonconvex FitzHugh–Nagumo parameter-estimation problem, the method matches the behavior of full-order inversion while cutting online cost. Reduced-coordinate inversion, tensor-based optimization, and suitable regularization further improve robustness when the problem is higher-dimensional, noisy, or strongly nonconvex. A reader cares because many scientific inversions are dominated by repeated expensive forward solves; compressing the observation map once and reusing it changes that cost structure.","feed_headline":"Tensor ROMs match full-order inverse solves at far lower cost","feed_subtitle":"TT compression of the parameter-to-observation map enables reduced-coordinate Gauss–Newton for heat and neural models.","key_machinery":"The central object is the tensor-train (TT) representation of the parameter-to-observation map, built by TT-SVD or TT-Cross. It replaces full-order dynamical solves with cheap TT contractions, enables reduced-coordinate Gauss–Newton assembly without forming the full observation-space Jacobian, and supports direct objective minimization over the discrete parameter grid.","core_discovery":"A tensor-train approximation of the parameter-to-observation map can be embedded in a regularized nonlinear least-squares inverse problem so that inversion proceeds in reduced coordinates, Gauss–Newton quantities are formed without the full Jacobian, and the recovered solutions reproduce full-order inversion at substantially reduced online cost. The same TT representation supports a stand-alone grid minimization or a warm start for Gauss–Newton, remaining effective under noise and nonconvexity when regularization and reduced coordinates are used.","pith_inferences":["Systems whose observation maps require high TT rank would lose the online advantage and may need adaptive or hierarchical tensor formats instead of fixed TT-SVD/TT-Cross.","Adaptive sampling of the parameter domain during offline TT construction could further cut the cost of building the reduced map for very high-dimensional parameters.","The reduced-coordinate residual landscape suggests a natural route to cheap sampling-based uncertainty quantification after the TT map is built.","The same TT compression of the parameter-to-observation map could accelerate other optimization-based inversions, such as material identification or inverse scattering, beyond the two examples studied."],"forward_implications":["Online inversion cost falls because each forward evaluation becomes a tensor contraction rather than a full dynamical-system solve.","Higher-dimensional parameter spaces remain tractable when the observation map stays low-rank in TT format.","Tensor-based grid minimization supplies a reliable initialization that steers Gauss–Newton away from poor local minima in nonconvex landscapes.","Regularized TROM inversion stays stable under measurement noise that would degrade unregularized full-order or reduced solves.","The same TT map can be reused for multiple inversions or different observation subsets without rebuilding the forward model."],"fun_headline_variants":["Tensor-train maps match full-order inversion at far lower online cost","TT compression enables reduced-coordinate Gauss-Newton without full Jacobians","TROM recovers inclusions and FHN parameters under noise via low-rank tensors","Reduced-order tensor least-squares matches full inversion for dynamical systems","Tensor grid minimization warms Gauss-Newton for nonconvex inverse problems"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The parameter-to-observation map of the target dynamical system must admit a sufficiently accurate low-rank tensor-train representation so that reduced-order approximation error does not spoil the recovered inverse solution.","fun_headline_variants_meta":{"raw":{"variants":["Tensor-train maps match full-order inversion at far lower online cost","TT compression enables reduced-coordinate Gauss-Newton without full Jacobians","TROM recovers inclusions and FHN parameters under noise via low-rank tensors","Reduced-order tensor least-squares matches full inversion for dynamical systems","Tensor grid minimization warms Gauss-Newton for nonconvex inverse problems"]},"model":"grok-4.5","effort":"low","cost_usd":0.00448,"raw_usage":{"total_tokens":1366,"prompt_tokens":833,"num_sources_used":0,"completion_tokens":96,"cost_in_usd_ticks":44800000,"prompt_tokens_details":{"text_tokens":833,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":437,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":833,"tokens_out":96,"duration_ms":3948,"temperature":1.0,"reasoning_tokens":437,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-15T04:43:00.904018+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Identify or construct a dynamical system whose parameter-to-observation map has high tensor-train rank; run TROM and full-order inversion on identical noisy data and check whether TROM recovers parameters that match the full-order solution within the paper’s reported error tolerances—systematic mismatch falsifies the claim that TROM reproduces full-order behavior.","supporting_citations":[],"review_version":1}