{"id":"83371610-e0dd-4ade-a167-4724f881acff","arxiv_id":"2608.06281","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":3,"one_line_summary":"Projected Hartree-Fock reproduces exact configuration-interaction electromagnetic moments and transitions reasonably well across selected sd- and pf-shell nuclei, with notable exceptions.","lead":"This paper compares a cheap nuclear-physics method, angular-momentum projected Hartree-Fock, with exact shell-model calculations for electric and magnetic observables. It finds mostly good agreement, including for odd-mass and odd-odd nuclei, but with several clear failures.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Selection of benchmark nuclei and absence of aggregate error metrics make the 'reasonably good agreement' claim anecdotal; a systematic sd-shell benchmark would settle it.","rationale":"The reader's conditional verdict is appropriate. The paper is an honest benchmark with disclosed failures, and the appendix gives a useful formal derivation of projected one-body densities with a number-operator consistency check. However, the headline generalization is not yet quantitatively grounded: the test set is hand-picked, 34S is excluded from the plotted results despite a factor-of-two B(E2) discrepancy, and no aggregate error metric is supplied. I would not reject: PHF-vs-FCI agreement is the right question, and the reported successes are plausible. I would keep the paper conditional, with a recommendation that the authors either supply a systematic aggregate benchmark or weaken the abstract's generalizing language. A secondary, easily fixed issue is in Section II.C: the quoted neutron bare spin g-factor, -3.2863 mu_N, is not the standard bare value (-3.8263 mu_N). Since PHF and FCI use the same operator, this does not invalidate the internal comparison, but it should be corrected before publication.","tokens_in":18085,"tokens_out":9329,"duration_ms":119239,"concrete_test":"Run a systematic sd-shell benchmark: all even-even, odd-A, and odd-odd sd-shell nuclides with FCI dimension below a fixed cutoff, using USDB and the same effective charges and g-factors as the paper. Match PHF and FCI states by maximum wave-function overlap rather than by J and energy order, and report, separately for E2 and M1 moments and B-values, the median relative deviation and the fraction of observables within 20% of FCI. If the aggregates are small across this full set, the selection concern is resolved; if only a few favorable nuclei drive the headline, the abstract's generalization is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that PHF reproduces FCI electromagnetic observables 'reasonably well,' including for many odd-A and odd-odd nuclides. For this to be robust, the benchmark set and the PHF-FCI state correspondence must be representative and unambiguous. Neither is currently established. Only about ten nuclei are presented, chosen without a stated selection criterion, and some striking failures live only in Appendix B: the 34S B(E2) values in Table XI are roughly a factor of two below FCI (e.g., 20.77 vs 37.09 e^2 fm^4 for 2+_1 -> 0+_1) yet 34S is not plotted in Section III. The odd-A/odd-odd conclusion rests on five nuclei with mixed outcomes: 30Al and 48V B(M1) and B(E2) have clear outliers, and 63Zn B(M1) does not agree well. Because no mean/median deviation or fraction-within-tolerance is reported, statements such as 'frequently' and 'overall better than expected' are not quantitatively testable. In dense odd-A/odd-odd spectra, states are labeled by J and energy ordering without overlap checks; a PHF state that is third in energy for a given J need not correspond to the FCI third state, so the apparent agreement or disagreement of transitions can be a labeling artifact. The load-bearing condition is that the selected cases and labels are unbiased, and the paper does not provide evidence for it.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper benchmarks angular-momentum projected after-variation Hartree-Fock (PHF) against full configuration interaction (FCI) for electric quadrupole (E2) and magnetic dipole (M1) moments and transitions in sd- and pf-shell nuclei. Both sets of calculations use the same model spaces, the same interactions (USDB and GX1A), and the same density-matrix pipeline, so the comparison isolates the PHF approximation error. The authors report reasonable agreement in many cases, including some odd-A and odd-odd nuclides, while explicitly acknowledging failures such as the 46Ti quadrupole observables and the 34S B(E2) values reported in Appendix B. An appendix derives the general formalism for one-body density matrices in multi-reference angular-momentum projection.","tokens_in":18388,"tokens_out":4663,"duration_ms":53816,"significance":"If the benchmarking were systematic, the paper would be a useful contribution: it demonstrates that a cheap mean-field-based method can reproduce several electromagnetic observables in a shell-model space, and it provides a detailed density-matrix formalism that will be useful for other applications such as dark-matter scattering calculations. The clean comparison setup, the absence of parameters fitted to the benchmarked quantities (effective charges, g-factors, and the oscillator parameter are standard literature values), and the explicit reporting of failures are all strengths. However, the current selection of nuclei and the lack of quantitative error metrics substantially limit the generality of the central claim. The methodological appendix is the strongest part of the paper; the benchmark conclusions need additional support before they can be regarded as a robust statement about PHF reliability.","major_comments":[{"comment":"The benchmark set is assembled without a stated selection criterion, and the central claim of 'reasonably good agreement' is not backed by quantitative aggregate measures. For example, the only data for 34S appear in Appendix B without discussion, and Table XI shows B(E2; 2+1 -> 0+1) = 20.77 e^2 fm^4 versus FCI 37.09 e^2 fm^4, a factor of 1.8 discrepancy; similarly, Section III.B reports 'poor agreement' for the 46Ti quadrupole observables. These failures are acknowledged but not integrated into any average deviation, median ratio, or fraction-within-tolerance statistic across the eleven nuclei. Without such metrics or a defined inclusion criterion, the qualitative conclusion is not falsifiable, and the phrase 'widest systematic such benchmarking' in Section III is not justified.","section":"Section III and Appendix B"},{"comment":"The correspondence between PHF and FCI states is not established for dense spectra. In Tables V, VII, and X and in the 48V and 63Zn discussions, states are labeled by total angular momentum and energy ordering, but no overlaps or wave-function similarity measures are reported. A PHF state that is the n-th state of a given J need not be the same physical state as the FCI n-th state, so comparisons of B(M1) and B(E2) values under this labeling can be meaningless. Because the paper's most surprising claim concerns odd-A and odd-odd nuclides, this state-assignment ambiguity is a load-bearing gap that should be addressed, for example by reporting overlaps or at least by restricting the comparison to states whose correspondence is supported by band structure or dominant configurations.","section":"Section III, odd-A and odd-odd subsections"},{"comment":"The claim that magnetic dipole 'transitions are overall better than expected' is not supported by the tabulated B(M1) data. In Table VII (25Mg), several B(M1) values differ from FCI by large factors, e.g., 13/2+1 -> 11/2+1: 0.01 vs 0.79 mu_N^2, and 5/2+2 -> 3/2+1: 0.01 vs 0.53 mu_N^2. The text in Section III.B states that B(M1) for 48V has 'several significant outliers' and that B(M1) for 63Zn 'do not agree well.' The abstract's wording thus appears to overstate the agreement for M1 transitions, even though M1 moments do agree well in several cases. This should be reworded to distinguish moments from transitions and to acknowledge the quantitative spread.","section":"Abstract and Section IV"}],"minor_comments":[{"comment":"The caption reads 'B(M2) transition strengths' but should read 'B(E2) transition strengths' for 64Cu.","section":"Figure 20"},{"comment":"Figure 18 appears to be an uncited duplicate of Figure 17(a); please remove it or reference it in the text.","section":"Figure 18"},{"comment":"There are several typos: 'transitinos' in Section III.A, 'Cartersian' in Section II.C, 'Clebsh-Gordan' in Appendix A, 'seperate' in the Table IV caption, 'valance' in the Table V caption, and 'Mev' in the 48V discussion in Section III.B.","section":"Throughout"},{"comment":"The axis labels in several scatter plots are small and difficult to read; increasing the font size or adjusting the layout would improve legibility.","section":"Figures 3, 5, 9, 12, 14, 19, 21"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of a nuclear structure journal and the methodological appendix is a useful contribution. The 'widest systematic benchmarking' phrase overstates the scope of the study; the authors should either substantially expand the benchmark set or soften the claim. The state-correspondence issue is significant and should be addressed before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Punchline: this is a solid, honest benchmarking paper that extends the same group's earlier PHF-vs-FCI spectra study to E2 and M1 observables, including odd-A and odd-odd nuclei, and the density-matrix appendix is a real contribution. The central claim—\"reasonably good agreement\"—is supported by the data as far as it goes, but the case selection is ad hoc and error reporting is qualitative, so the strength of the claim is not as sharply established as it could be.\n\nWhat is actually new: the systematic head-to-head comparison in identical model spaces with the same interaction and the same density-matrix pipeline, covering nine nuclei across sd and pf shells. Previous PHF electromagnetic benchmarks exist (Gunye-Warke, Baye-Descouvemont), but I haven't seen this breadth of odd-A and odd-odd cases. The appendix with multi-reference density-matrix formalism, including overlap checks in Eqs. (A20-A21), is useful and will be cited.\n\nWhat the paper does well: it is scrupulously fair. The failures are disclosed—46Ti quadrupole moments and B(E2)s are far off, and 34S B(E2) values in Table XI are nearly a factor of two below FCI. The workflow diagram is clear, and using the same code for both FCI and PHF observables removes a whole class of implementation artifacts. The claim that M1 observables are \"better than expected\" is presented with enough data to be plausible.\n\nSoft spots, in order: (1) The benchmark set is hand-picked. The paper says \"select cases\" but gives no criterion, and 34S—a striking failure—is only in an appendix, not plotted in the main text. With only nine nuclei, the generalization to \"PHF is a reliable starting point\" is anecdotal. A table of mean/median deviations or fraction-within-tolerance across a defined set would nail it. (2) In dense odd-A and odd-odd spectra, states are matched by J and energy order without overlap checks. If the PHF and FCI states are not the same physical states, some of the apparent agreement in transition strengths could be a labeling coincidence. This is a real concern, though it doesn't undercut the moments, which are state-by-state. (3) No code or data release, which is a missed opportunity for a benchmarking paper.\n\nOverall: the core comparison is sound, the methodology is honest, and the appendix has lasting value. The claim is appropriately qualified, but the selection and metric issues should be addressed. I'd send this to a serious referee and ask for more systematic case selection or aggregate statistics, plus overlap checks in odd-A/odd-odd labeling. It deserves publication after a moderate revision.","headline":"A clean and honest PHF-vs-FCI benchmark for E2/M1 observables, with a useful density-matrix appendix; the hand-picked cases and qualitative error metrics keep the 'reasonably good agreement' claim from being fully quantitative.","tokens_in":18896,"tokens_out":2402,"would_cite":true,"duration_ms":26328,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Angular-momentum projected Hartree-Fock reproduces full configuration-interaction electromagnetic observables reasonably well across selected sd- and pf-shell nuclei, including odd-A and odd-odd cases, though a few clear failures remain.","keywords":["angular-momentum projection","projected Hartree-Fock","shell model","configuration interaction benchmark","E2 transitions","M1 transitions","electromagnetic moments","sd/pf valence spaces"],"falsifier":"Compute PHF and FCI electromagnetic observables for all sd-shell and pf-shell nuclei (even-even, odd-A, and odd-odd) with the same interactions and operator conventions, and compare average relative deviations in B(E2) and B(M1); if the mean deviation over the full set is much larger than the scatter in the cases plotted here, the claim of reasonably good agreement would be refuted. A single sharper check: if the near-factor-of-two error in 34S B(E2) values recurs across a wide sample, the successes in the other nuclei are not representative.","tokens_in":17899,"feed_emoji":"⚛️","tokens_out":7950,"duration_ms":87225,"temperature":0.7,"pith_summary":"Angular-momentum projected Hartree-Fock (PHF) is a deliberately simple approximation: a single Slater determinant minimized by Hartree-Fock, then projected onto good angular momentum after variation. This paper asks whether such a cheap state can reproduce the electric quadrupole (E2) and magnetic dipole (M1) moments and transition strengths of full configuration-interaction (FCI) shell-model calculations. Benchmarking selected nuclei in the sd and pf shells with identical model spaces and interactions, the authors find reasonably good agreement for many even-even, odd-A, and odd-odd nuclei, with M1 observables agreeing better than expected. The agreement is qualified: quadrupole observables in 46Ti and B(E2) values in 34S are poorly reproduced. The point matters because PHF is a stepping stone for more elaborate many-body methods, so knowing what its bare densities get right is useful.","feed_headline":"Simple projected Hartree-Fock matches shell-model E2 and M1 benchmarks","feed_subtitle":"Cheap mean-field states reproduce full shell-model E2 and M1 results in many odd-A and odd-odd nuclei.","key_machinery":"The carrying mechanism is angular-momentum projected-after-variation Hartree-Fock (PHF): one or more real Slater determinants obtained by unconstrained Hartree-Fock minimization, projected onto good total angular momentum by solving a small linear-algebra problem rather than by quadrature. From the projected states the authors construct one-body density matrices for arbitrary angular-momentum rank, then fold in single-particle reduced matrix elements, with effective charges and bare g-factors, to obtain moments and B-values. The same density-matrix pipeline is applied to FCI wave functions, so the two methods differ only in the densities, not in the operator treatment.","core_discovery":"The paper's central claim is that PHF, with no adjustment beyond the standard effective charges and bare g-factors, reproduces FCI electromagnetic observables well enough to serve as a simple alternative and as a foundation for more sophisticated methods. Using the same shell-model Hamiltonian and basis for both PHF and FCI, the authors extract one-body density matrices from angular-momentum-projected states and compute E2 and M1 moments and transitions through reduced matrix elements. They report that M1 moments are frequently well reproduced, including in odd-A and odd-odd nuclei, and that E2 observables are usually but not always reliable, with the exceptions of 46Ti quadrupole moments and transitions and 34S B(E2) values. The paper also notes that including multiple Hartree-Fock minima improves agreement in shape-coexisting cases, consistent with earlier spectral benchmarks.","pith_inferences":["Because the benchmark set is small and selected without a stated sampling rule, the 'reasonably good agreement' conclusion is likely to be read as a proof of concept rather than a statistical statement; a comprehensive sd/pf scan reporting mean absolute deviations would either confirm the pattern or reveal that the selected nuclei are unusually favorable.","The 46Ti and 34S failures suggest that quadrupole agreement is not guaranteed by deformation alone; a natural testable hypothesis is that PHF fails when the FCI state has meaningful configuration mixing beyond a single intrinsic shape, and that adding more reference states or a generator-coordinate step restores agreement.","The general one-body density formalism in the appendix could be reused for other rank-one and rank-two operators, such as those entering neutrino responses or dark-matter scattering, without re-deriving the projection machinery."],"forward_implications":["PHF can stand in for FCI for electromagnetic observables in medium-light nuclei when full diagonalization is impractical, at least for the kinds of states tested here.","The reasonable M1 agreement indicates that simple projected mean-field states carry enough spin and orbital structure to be a useful starting point for magnetic observables, not just spectra.","When shape coexistence is present, using multiple Hartree-Fock minima rather than a single one is a cheap way to improve both spectra and electromagnetic observables.","The documented failures in 46Ti and 34S mean PHF should be validated against FCI or experiment before being trusted for quadrupole observables in specific nuclei."],"supporting_citations":[{"why":"Previous benchmark of PHF excitation spectra against FCI; this paper extends that comparison to electromagnetic observables and reuses its method and earlier conclusions.","marker":"[16]"},{"why":"Configuration-interaction diagonalization code that generates the FCI wave functions and densities used as the benchmark.","marker":"[18, 19]"},{"why":"Linear-algebra method for angular-momentum projection used to construct PHF states without quadrature.","marker":"[22]"},{"why":"Source for the effective charges used in the E2 operator, so the PHF-FCI comparison is made with standard operator inputs.","marker":"[24, 25]"},{"why":"USDB interaction defining the sd-shell Hamiltonian used for all sd benchmarks.","marker":"[32]"},{"why":"Source of the modified G-matrix GX1A interaction used in the pf-shell calculations.","marker":"[33–35]"}],"fun_headline_variants":["Projected HF reproduces shell-model E2 and M1 benchmarks broadly","Simple projected Hartree-Fock matches M1 moments in odd-A nuclei","Mean-field PHF matches full-CI electromagnetic observables","PHF electromagnetic results rival full shell-model calculations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion that PHF is broadly reliable rests on the assumption that the select sd- and pf-shell nuclei are representative, since no systematic criterion or aggregate error statistic over the full space is given.","fun_headline_variants_meta":{"raw":{"variants":["Projected HF reproduces shell-model E2 and M1 benchmarks broadly","Simple projected Hartree-Fock matches M1 moments in odd-A nuclei","Mean-field PHF matches full-CI electromagnetic observables","PHF electromagnetic results rival full shell-model calculations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000583,"raw_usage":{"total_tokens":2711,"prompt_tokens":878,"completion_tokens":1833,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":494,"completion_tokens_details":{"reasoning_tokens":1762}},"tokens_in":494,"tokens_out":1833,"duration_ms":17162,"temperature":1.0,"reasoning_tokens":1762,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T10:33:10.824285+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute PHF and FCI electromagnetic observables for all sd-shell and pf-shell nuclei (even-even, odd-A, and odd-odd) with the same interactions and operator conventions, and compare average relative deviations in B(E2) and B(M1); if the mean deviation over the full set is much larger than the scatter in the cases plotted here, the claim of reasonably good agreement would be refuted. A single sharper check: if the near-factor-of-two error in 34S B(E2) values recurs across a wide sample, the successes in the other nuclei are not representative.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"USDB interaction defining the sd-shell Hamiltonian used for all sd benchmarks."},{"cited_title":"Caurier, G","cited_arxiv_id":null,"evidence_quote":"Previous benchmark of PHF excitation spectra against FCI; this paper extends that comparison to electromagnetic observables and reuses its method and earlier conclusions."},{"cited_title":"Hagen, S","cited_arxiv_id":null,"evidence_quote":"Linear-algebra method for angular-momentum projection used to construct PHF states without quadrature."}],"review_version":1}