{"id":"1cb8fda0-7bfe-4cce-aea5-5c47795e5ce6","arxiv_id":"2502.10053","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":1,"one_line_summary":"A covariant Moore-Penrose algorithm for complex degenerate metrics is formulated, but its uniqueness and torsion interpretation depend on an arbitrary auxiliary metric and the central proof is incomplete.","lead":"This paper introduces a coordinate-friendly way to compute inverses of broken, complex-valued spacetime metrics, and uses it to study black holes and cosmology. The method's results depend on a helper metric the authors choose by hand, and the main proof has gaps.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Appendix A's Cholesky construction does not satisfy the covariant adjoint conditions (51c,d) for generic complex metrics; (52) also omits complex conjugation, so the claimed uniqueness and tensoriality of the pseudoinverse are unestablished.","rationale":"The reader's weakest assumption—that the Appendix A construction satisfies the ζ-twisted adjoint conditions—is exactly where the paper fails. Our independent computation with a 2×2 non-commuting example shows the Cholesky-translated standard Moore-Penrose inverse does not satisfy (51c). The cause is an algebraic mismatch: the standard MP inverse is defined with respect to the ordinary Hermitian inner product, and the Cholesky factor does not commute with the rank-1 projector Q = g_0 \\tilde{g}_0 unless g and ζ commute. In addition, the component equations (52c,d) omit the complex conjugation present in the definitions (47)-(48), so even the stated identities are not the covariant ones. Because existence and uniqueness of \\tilde{g} underlie the connection (55), torsion (62), and curvature tensors, the central claim of a covariant extension of GR to degenerate complex metrics is unsubstantiated. The Schwarzschild and Reissner-Nordström applications exploit the commuting special case and do not establish the general algorithm. The geodesic completeness proof (Sec. II A 2) is a plausible side result but independent of the pseudoinverse construction. Thus the rejection stands.","tokens_in":79,"tokens_out":28544,"duration_ms":792717,"concrete_test":"Run the Appendix A recipe symbolically on the 2×2 counterexample: g = [[1,i],[i,-1]], ζ = diag(2,1), B = diag(√2,1). Compute g_0 = B^{-1} g B^{-1}, its Moore-Penrose inverse \\tilde{g}_0, and \\tilde{g} = B^{-1} \\tilde{g}_0 B^{-1}. Then verify the ζ-twisted adjoint identities (51c,d) by computing M = g\\tilde{g} and N = \\tilde{g}g and checking whether M† = ζ^{-1} \\bar{M}^T ζ equals M and †N = ζ^{-1} \\bar{N}^T ζ equals N. If (as our manual calculation indicates) equality fails for M, the Appendix A existence theorem is false as stated; repeating with ζ = [[2,1],[1,2]] (non-diagonal) would further confirm the failure is generic, not an artifact of a diagonal ζ.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central existence/uniqueness theorem (Appendix A) is invalid. The covariant adjoint (47)-(48) includes complex conjugation, so the component identities (52c,d) as printed—without overbars—are not equivalent to (51c,d) for complex metrics. More seriously, the Cholesky construction fails the intended conditions. With ζ = B^T B, the paper defines g_0 = B^{-T} g B^{-1}, obtains its standard Moore-Penrose inverse \\tilde{g}_0, and sets \\tilde{g} = B^{-1} \\tilde{g}_0 B^{-T}. Then M := g\\tilde{g} = B^T Q B^{-T} with Q = g_0 \\tilde{g}_0. The standard MP conditions only imply Q is Hermitian. The ζ-twisted adjoint satisfies M† = ζ^{-1} \\bar{M}^T ζ = B^{-1}(B^{-T}B^{-1}) \\bar{Q}^T (B B^T) B, which is not generally equal to M. Concrete counterexample: g = [[1,i],[i,-1]], ζ = diag(2,1), B = diag(√2,1). Computation yields \\tilde{g} = [[1/9,-2i/9],[-2i/9,-4/9]], so M = g\\tilde{g} = [[1/3,-2i/3],[i/3,2/3]], but M† = [[1/3,-i/6],[4i/3,2/3]] ≠ M. Thus (51c) fails. Consequently the connection (55), torsion (62), and curvature (63)-(71) are not genuine tensors, and the central claim of a covariant extension of the Moore-Penrose algorithm is unsupported for generic complex degenerate metrics.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a covariant extension of the Moore-Penrose pseudoinverse for constant-rank symmetric complex degenerate metrics, replacing the ordinary adjoint by a ζ-twisted adjoint defined via an auxiliary Riemannian metric ζ. The authors claim that the resulting pseudoinverse is unique, that the connection (55) and curvature tensors (63)–(71) are genuine tensors, and that metric degeneracy is encoded in a torsion tensor. These tools are applied to complexified Schwarzschild, Reissner–Nordström, Reissner–Nordström–de Sitter, and FLRW metrics, and a new notion of extremal curves is introduced. The paper also contains a self-contained proof of geodesic completeness of the Euclidean section of Schwarzschild (Sec. II A 2).","tokens_in":33522,"tokens_out":8856,"duration_ms":80825,"significance":"If the central existence/uniqueness theorem were correct, the framework would provide a principled way to handle nowhere-invertible complex metrics and could be relevant to quantum-gravity path integrals and signature-change models. The paper is transparent about the role of the auxiliary metric ζ and includes a nontrivial original proof of Euclidean Schwarzschild geodesic completeness. However, the central construction fails for generic complex metrics: the component equations (52) are not equivalent to the ζ-twisted adjoint conditions (51), and a simple counterexample in the flat case shows that the Cholesky-translated pseudoinverse does not satisfy (51c). Since the connection, torsion, and curvature tensors all rely on this construction, the main claim of the paper is not established. The ζ-dependence of the torsion—acknowledged by the authors themselves—further weakens the physical interpretation.","major_comments":[{"comment":"The component equations (52) do not reproduce the covariant adjoint conditions (51) for complex metrics. The right and left dagger operations (47) and (48) include complex conjugation of the tensor components, but the printed component identities (52c) and (52d) contain no overbars. For a complex metric, (52c) is therefore not equivalent to (g˜g)† = g˜g, and the claim in Sec. III A that (52) is 'equivalently' (51) is incorrect. The uniqueness proof in Appendix A is carried out for the wrong set of equations in the complex case.","section":"Appendix A; Sec. III A, Eqs. (51)-(52)"},{"comment":"The Cholesky-based construction in Appendix A does not yield a tensor satisfying the ζ-twisted adjoint conditions. With ζ = B^T B, the paper sets g_0 = B^{-T} g B^{-1} and tilde g = B^{-1} tilde g_0 B^{-T}, where tilde g_0 is the standard Moore-Penrose inverse. But the standard conditions only imply that g_0 tilde g_0 is Hermitian, not that g tilde g is ζ-self-adjoint. For a concrete failing example, take g = [[1,i],[i,-1]] and ζ = diag(2,1), so B = diag(sqrt(2),1). The construction gives tilde g = [[1/9,-2i/9],[-2i/9,-4/9]], and direct computation yields g tilde g = [[1/3,-2i/3],[i/3,2/3]], whereas the ζ-twisted adjoint of g tilde g evaluates to [[1/3,-i/6],[4i/3,2/3]], which is not equal to g tilde g. Hence condition (51c) fails even locally in flat space, so the existence and uniqueness claim of Sec. III A is false as stated; consequently the tensorial character of the connection (55), the torsion (62), and the curvature tensors (63)–(71) rests on an invalid premise.","section":"Appendix A; Sec. III A"},{"comment":"The torsion tensor (62) and the connection (55) depend on the arbitrary auxiliary metric ζ_ab. Condition (B1c) in Appendix B is introduced as a normalization without a geometric or physical justification, and the paper itself states in Sec. V that no natural choice of ζ is evident and that the dependence of physical predictions on ζ deserves separate study. As a result, the claim that metric degeneracy is 'geometrically represented' by a torsion tensor is not gauge-invariant: different choices of ζ can change the torsion even for the same degenerate metric, so the physical content of the framework is not fully determined by the metric alone.","section":"Sec. III B, Eq. (62); Appendix B; Sec. V"}],"minor_comments":[{"comment":"The metric is described as a '(2,0)-tensor field' but also as 'a global section of T*M⊗T*M'; a (2,0) tensor is a section of TM⊗TM, and later examples write gab with lower indices. Please make the index convention consistent throughout.","section":"Sec. III A"},{"comment":"The proof of geodesic completeness refers to Eq. (27) before that equation is derived; moving the derivation of the radial equation ahead of the initial-condition discussion would improve readability.","section":"Sec. II A 2"},{"comment":"The index placement in the dagger operations (47) and (48) is unclear: for a field X^a_b, the right-hand side ζ^{ac} X^d_c ζ_{db} has the same index type as X, so the transposition of X is not displayed. Writing the complex transpose explicitly, for example ζ^{ac} \\overline{X^c_d} ζ_{db}, would remove the ambiguity.","section":"Sec. III A, Eqs. (47)-(48)"}],"recommendation":"reject","confidential_remarks":"The central theorem is invalid for complex metrics, so this is a clear reject. The original proof of Euclidean Schwarzschild geodesic completeness in Sec. II A 2 is a useful isolated result, but it does not compensate for the failure of the main construction. The paper is appropriate for gr-qc in principle, but the main claim must be fixed before resubmission."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the central algorithm doesn't work as stated. Appendix A's Cholesky construction does not satisfy the ζ-twisted adjoint conditions (51c,d) for generic complex metrics. The component equations (52c,d) drop the complex conjugations that the adjoint (47) requires, and a concrete 2x2 example—g = [[1,i],[i,-1]], ζ = diag(2,1), B = diag(√2,1)—shows the constructed pseudoinverse violates (51c). So the claimed uniqueness and tensoriality of the pseudoinverse metric are not established, and the connection, torsion, and curvature built on it are not genuine tensors for the general framework the paper advertises.\n\nWhat is genuinely good: the paper is clearly written and the ambition is sensible. The geodesic completeness proof for Euclidean Schwarzschild in Kruskal-Szekeres coordinates (Sec. II A 2) looks plausible and is an original side result. The FLRW analysis—where the torsion tensor vanishes exactly when the scale factor is holomorphic (Sec. III D 3, Eq. (94))—is a neat observation, as is the explicit acknowledgment in Sec. V that the auxiliary metric ζ_ab is an arbitrary input with no natural choice yet identified. That admission is honest, but it also reinforces the circularity problem: all physical quantities depend on ζ, and the paper offers no physical principle to fix it.\n\nThe soft spot is load-bearing, not cosmetic. The Schwarzschild example works only because the degenerate (T,t) block sees ζ = identity; with a non-identity ζ the standard Moore-Penrose inverse fails the covariant adjoint condition even for that simple block. The general existence theorem in Appendix A is therefore false as stated, and the torsion interpretation is an artifact of an unverified construction.\n\nWho is this for? Readers interested in complex degenerate metrics in quantum gravity or signature change might find the geodesic completeness proof and the holomorphy/torsion connection useful. Anyone seeking a working covariant pseudoinverse algorithm will not find it here.\n\nRecommendation: this deserves a serious referee, because the flaw is subtle and the topic is important, but the paper should not be published as is. The side results could be salvaged; the central construction needs a corrected adjoint definition and a valid existence proof.","headline":"The advertised covariant Moore-Penrose construction fails its own adjoint conditions for generic complex metrics, so the central uniqueness/tensoriality claims are unsupported; the geodesic completeness proof and the FLRW holomorphy observation are worth preserving.","tokens_in":34116,"tokens_out":7109,"would_cite":false,"duration_ms":64959,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["15A09","53B20","83C05","83F05"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper constructs a covariant Moore-Penrose pseudoinverse that gives complex degenerate metrics unique, well-defined curvature tensors.","keywords":["covariant Moore-Penrose pseudoinverse","degenerate metric","complex metrics","torsion tensor","complexified Schwarzschild spacetime","FLRW cosmology","geodesics of degenerate metrics","general relativity"],"falsifier":"Take a small constant-rank complex symmetric matrix $g$ that does not commute with a chosen positive-definite $\\zeta$, construct the candidate $\\tilde{g} = B^{-1}({}^t B^{-1} g B^{-1})^{+} {}^t B^{-1}$ from the Cholesky factor $B$ of $\\zeta$, and test whether the component identities (52c) and (52d) hold exactly; a single failure would falsify the general existence theorem as stated.","tokens_in":32916,"feed_emoji":"🌌","tokens_out":6348,"duration_ms":62351,"temperature":0.7,"pith_summary":"The paper tries to establish that general relativity can be extended to metrics that are complex and nowhere invertible, provided one replaces the inverse metric by a pseudoinverse selected by coordinate-covariant conditions. The central object is a symmetric tensor $\\tilde{g}^{ab}$ defined by four tensor identities that generalize the Moore-Penrose axioms, with the adjoint evaluated using a fixed background Riemannian metric $\\zeta_{ab}$. The authors argue that this pseudoinverse is unique, transforms as a tensor, and yields a metric-compatible connection whose torsion measures the degeneracy. If correct, this gives well-defined curvature tensors, Einstein equations, and geodesic notions for non-invertible complex spacetimes, covering complexified black holes and FLRW cosmology.","feed_headline":"A covariant inverse makes degenerate metrics usable in relativity","feed_subtitle":"Unique pseudoinverse gives non-invertible complex metrics a true connection and curvature, with black-hole and cosmological applications.","key_machinery":"The machinery is a covariant dagger adjoint built from a fixed Riemannian metric $\\zeta$: for tensor components, the adjoint uses $\\zeta$ to raise and lower indices after complex conjugation, so the standard Moore-Penrose conditions become tensor equations. Existence and uniqueness are obtained by writing $\\zeta = {}^t B B$ (the Cholesky decomposition), pulling the degenerate metric $g$ back to $g_0 = {}^t B^{-1} g B^{-1}$, taking its ordinary Moore-Penrose inverse, and pushing the result back. This object carries the argument because it makes the connection (55), the torsion (62), and the curvature invariants well-defined coordinate-invariant quantities.","core_discovery":"For any symmetric complex degenerate metric of constant rank, the covariant Moore-Penrose conditions (51) have a unique solution $\\tilde{g}$ with respect to any fixed Riemannian metric $\\zeta$; the proof proceeds by Cholesky decomposition of $\\zeta$ in local charts and gluing. This makes $\\tilde{g}$ a true tensor, so the connection (55), Riemann tensor (63), Ricci tensor (67), and Einstein tensor (71) are genuine tensors whose coordinate transformation is unproblematic. The connection is $g$-compatible, but $\\tilde{g}$ is generally not covariantly constant, and the antisymmetric part of the connection yields a torsion tensor (62) that vanishes only for maximal-rank or highly symmetric metrics. Applications show that complexified Schwarzschild and Reissner-Nordström metrics satisfy the vacuum or Einstein-Maxwell equations with the same Kretschmann scalar as their ordinary counterparts, while the FLRW model yields complex Friedmann equations and interprets holomorphy of the scale factor as the condition of vanishing torsion.","pith_inferences":["Editorial inference: A direct numerical check of the four covariant identities for arbitrary constant-rank complex symmetric matrices, using a non-commuting positive-definite $\\zeta$, would settle the existence claim independently of the Appendix A proof.","Editorial inference: If the construction extends, every degenerate symmetric $(2,0)$ tensor field, not just metrics, could be inverted covariantly, giving a general calculus for degenerate tensors in any dimension.","Editorial inference: The dependence on the auxiliary metric $\\zeta$ is a genuine freedom, so physical predictions such as torsion and conservation laws may shift under different choices of $\\zeta$; selecting $\\zeta$ naturally becomes the next question the framework raises.","Editorial inference: The proof that Euclidean Schwarzschild is geodesically complete while Lorentzian extremals can reach $r=0$ suggests that whether a complexified spacetime is singular may depend on which real section the geodesic is confined to."],"forward_implications":["Complexified Schwarzschild and Reissner-Nordström metrics can be handled as genuine spacetimes with $\\tilde{g}$, and their Kretschmann scalars coincide with the real-valued solutions, locating the singularity at $r=0$.","For the complex FLRW metric, the covariant pseudoinverse gives complex Friedmann equations; the conservation law $\\nabla_a T^{ab}=0$ holds exactly when the scale factor is a holomorphic function of $\\tau = T + i t$.","In general degenerate geometries the Einstein tensor no longer satisfies the ordinary conservation law; the modified identity (73) contains extra terms involving $\\nabla_a \\tilde{g}^{bc}$ and the torsion.","When the metric has maximal rank or high symmetry, such as Schwarzschild, Reissner-Nordström, and Reissner-Nordström-de Sitter, the torsion vanishes and the connection reduces to the pseudoinverse-mediated Christoffel form (77).","Extremal curves, called contravariant autoparallels, provide the degenerate-metric analogue of geodesics; in the Schwarzschild model they run entirely inside either the Euclidean or the Lorentzian section."],"supporting_citations":[{"why":"Supplies the classical Moore-Penrose existence, uniqueness, and smoothness results for constant-rank matrices that the covariant construction extends.","marker":"[65]"},{"why":"Provides the original pseudoinverse conditions that the paper reformulates in covariant form.","marker":"[64]"},{"why":"Gives the Cholesky decomposition theorem used to pull back the metric and its pseudoinverse.","marker":"[98]"},{"why":"Supplies the differentiability of the Cholesky factor needed to prove that the local pseudoinverse is smooth.","marker":"[100]"},{"why":"Provides the constant-rank smoothness formula for the classical Moore-Penrose inverse used in the existence proof.","marker":"[97]"},{"why":"Establishes the spacetime-defect framework and the geodesic equation for degenerate metrics that the extremal-curve discussion builds on.","marker":"[49]"},{"why":"Provides the complexified black-hole geometry and change-of-signature setup used as the main application.","marker":"[29]"},{"why":"Shows that degenerate tetrads allow nonvanishing torsion of geometric origin, supporting the paper's torsion interpretation.","marker":"[37]"},{"why":"Supplies the Euclidean Reissner-Nordström solution and particle dynamics that the complexified Reissner-Nordström application extends.","marker":"[78]"}],"fun_headline_variants":["Covariant Moore-Penrose: unique inverse for complex degenerate metrics","Torsion as the geometric marker of degenerate metrics","New covariant inverse makes complex degenerate metrics usable in relativity","Unique pseudoinverse metric enables black hole and cosmology models","Complex metrics without inverse? Covariant method provides solution"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction's uniqueness and tensoriality rest on Appendix A's claim that pulling back through the Cholesky factor of the fixed metric $\\zeta$ always produces a tensor satisfying the covariant adjoint conditions; if this does not hold for complex metrics, the pseudoinverse metric and every derived object lose their coordinate invariance.","fun_headline_variants_meta":{"raw":{"variants":["Covariant Moore-Penrose: unique inverse for complex degenerate metrics","Torsion as the geometric marker of degenerate metrics","New covariant inverse makes complex degenerate metrics usable in relativity","Unique pseudoinverse metric enables black hole and cosmology models","Complex metrics without inverse? Covariant method provides solution"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000282,"raw_usage":{"total_tokens":1649,"prompt_tokens":910,"completion_tokens":739,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":526,"completion_tokens_details":{"reasoning_tokens":657}},"tokens_in":526,"tokens_out":739,"duration_ms":7577,"temperature":1.0,"reasoning_tokens":657,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T19:37:55.383862+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a small constant-rank complex symmetric matrix $g$ that does not commute with a chosen positive-definite $\\zeta$, construct the candidate $\\tilde{g} = B^{-1}({}^t B^{-1} g B^{-1})^{+} {}^t B^{-1}$ from the Cholesky factor $B$ of $\\zeta$, and test whether the component identities (52c) and (52d) hold exactly; a single failure would falsify the general existence theorem as stated.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the original pseudoinverse conditions that the paper reformulates in covariant form."},{"cited_title":"Saerkkae, Bayesian Filtering and Smoothing , Institute of Mathematical Statistics Text- books (Cambridge University Press, 2013)","cited_arxiv_id":null,"evidence_quote":"Gives the Cholesky decomposition theorem used to pull back the metric and its pseudoinverse."}],"review_version":1}