{"id":"5ea4a5aa-9a8a-4035-88a1-43387733204f","arxiv_id":"2501.04214","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Any ghost-free spin-2 or spin-3 TDiff model is equivalent, via source and field redefinitions, to the Maxwell-like model, which matches the string tensionless doublet action.","lead":"This paper shows that all ghost-free spin-2 and spin-3 models with transverse diffeomorphism symmetry are physically equivalent to their simplest 'Maxwell-like' versions. It also connects these models to the tensionless limit of open bosonic string field theory via Stueckelberg fields.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The spin-3 gauge-invariant amplitude (22), on which the Section 5.2 equivalence proof rests, is asserted without derivation; its exact fD+2 coefficient and spin-1 source identification are not checked.","rationale":"The reader's verdict is CONDITIONAL and already identifies the same weakest assumption: the spin-3 amplitude (22) is the load-bearing unshown step. I agree. The spin-2 chain (6), (58)-(71) is explicit and algebraically checkable; the spin-3 equivalence chain in Section 5.2 hinges on (22). If (22) is correct, the source-redefinition argument goes through: Eq. (84) forces B = A^2 via fD+2(A,B) = (A-1)^2, and (86)-(89) ensure that any point with fD+2 > 0 can be r-shifted to (A,A^2). The treatment of the exceptional points a=1 and a=2/(D+2) is plausible but compressed. I see no actual algebraic error in the displayed steps, and there is independent internal support from the diagonalized forms (47) and (55) and from the claimed Mathematica check of the nonlocal redefinitions (98)-(101), so the concern is addressable rather than disqualifying. Since the reader already conditioned acceptance on this point, I recommend no change to the verdict: it should remain CONDITIONAL, with the requested check being a full independent derivation of (22).","tokens_in":19933,"tokens_out":12654,"duration_ms":130935,"concrete_test":"Independently rederive (22) by direct inversion of the gauge-fixed kinetic operator for S(a,b,cS) using the gauge-fixing term (21), without relying on the projection-operator shortcut. Concretely: use computer algebra (xAct/Mathematica) to compute the saturated two-point amplitude for generic sources satisfying (17) in D=4 and then general D, at symbolic a,b,D; verify that the spin-1 pole coefficient is exactly 3/[D fD+2(a,b)] and that the conserved source is (24). Also check the a=1 and a=2/(D+2) limiting cases separately. If the coefficient or source differs, the Section 5.2 equivalence conditions and the ghost-free region would need revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that all ghost-free spin-3 TDiff models are equivalent to the Maxwell-like model depends on the two-point amplitude (22). Section 5.2 uses (22) to impose equality between the Maxwell-like amplitude with s-shifted sources and the general S(A,B,cS) amplitude; Eq. (84) and the final condition B = A^2 follow only if the spin-1 residue is exactly 3/(D fD+2)|JtildeT|^2 with JtildeT = (2-a(D+2))Jmu + (a-1)Tmu. The text states that this was obtained with the rank-3 projection operators listed in the appendix, but the actual projection/contraction leading to (22) is not shown. The special cases a=1 and a=2/(D+2) are likewise handled by assertion ('basically the same expression'), not by a displayed computation. Because the same amplitude fixes the ghost-free boundary fD+2 > 0, an error in the coefficient or in the spin-1 source would propagate directly into the main equivalence theorem. This is a missing-support issue, not an observed contradiction, but it is the most load-bearing unverified step in the paper.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper investigates a two-parameter family of massless spin-2 and spin-3 theories invariant under (generalized) transverse diffeomorphisms. For spin-3, the most general second-order action with TrTDiff symmetry is reduced to a two-parameter family by requiring invariance under a generalized TDiff transformation, giving c = c_S. The paper computes the gauge-invariant two-point amplitudes, identifies the ghost-free region f_D > 0 (spin-2) and f_{D+2} > 0 (spin-3), and shows how the models connect to the doublet action from the tensionless limit of open bosonic string field theory via nonlocal Stueckelberg-type field redefinitions. The central new result is a duality: within the ghost-free region, every model can be mapped to the Maxwell-like model (a,b) = (0,0) through local r-shifts and source s-shifts, with equality of the two-point amplitudes. The spin-2 part is largely a review; the spin-3 part contains the principal new claims.","tokens_in":20160,"tokens_out":12876,"duration_ms":114530,"significance":"If correct, the result is significant: it establishes that the free physical content of ghost-free TDiff models of spin-2 and spin-3 is independent of the parameters (a,b) within the ghost-free region, unifying them with the Maxwell-like models and with the string-theory doublet action. The proof is algebraic and largely explicit, and the paper provides several cross-checks: the diagonalized Lagrangians (37), (47), (55) corroborate the amplitude results, and the nonlocal field redefinition in Sec. 5.3 is reported to have been checked with Mathematica/xAct. The paper is also candid about the limitations of nonlocal redefinitions. However, the spin-3 two-point amplitude (22), on which the spin-3 equivalence proof rests, is asserted without a displayed derivation; this is the main missing support for an otherwise coherent and interesting central claim.","major_comments":[{"comment":"The gauge-invariant two-point amplitude for the spin-3 model is stated without derivation. The text says it was obtained using the rank-3 projection operators of the appendix, but the actual projection-operator contraction is not shown. This is load-bearing: the coefficient 3/(D f_{D+2}) of the spin-1 residue and the definition of f_{D+2} determine the ghost-free region f_{D+2} > 0, and the equivalence proof of Sec. 5.2 uses (22) directly to derive the condition B = A^2. Please include the derivation of (22), at least for the generic case a ≠ 1, a ≠ 2/(D+2), and state explicitly how the special cases follow.","section":"Sec. 3.1, Eq. (22)"},{"comment":"The statements that the a = 1 case and the a = 2/(D+2) case lead to \"basically the same expression\" (22) are made without any displayed computation. Since these cases are part of the parameter space and because the r-shift roots used in Sec. 5.2 must be shown not to land on A = 1 (or on the fixed point A = 2/(D+2) when treating the generic family), the text should provide the explicit projection-operator result or a clear argument that the special cases are obtained as limits of the generic derivation.","section":"Sec. 3.1 and Sec. 3.2"},{"comment":"The alternative proof of equivalence via nonlocal field redefinitions relies on the statement \"We have explicitly checked by means of the software Mathematica with help of the package xAct\" that (98) with (100) and (101) maps S(A,B,c_S) to the Maxwell-like model when B = A^2. This computation is not shown and no notebook is provided, so the claim is not independently verifiable from the text. Please either include the computation in an appendix, provide the explicit resulting action after substitution, or make the code available as supplementary material.","section":"Sec. 5.3"}],"minor_comments":[{"comment":"The notation \"3/D f_{D+2}\" is ambiguous; it should read 3/(D f_{D+2}) to avoid confusion.","section":"Sec. 3.1, Eq. (22)"},{"comment":"The term \"6 χ (∂·∂·φ)\" is ambiguous because χ is a vector field in these equations; please use explicit index notation such as 6 χ_μ (∂·∂·φ)^μ.","section":"Sec. 4.2, Eq. (44) and Eq. (48)"},{"comment":"In the second term, \"s2 ∂_{(μ}∂_ν h_{ρ)}/□\", the index symmetrization should be clarified; write it as s2 ∂_{(μ}∂_ν h_{ρ)}/□ with the understanding that h_ρ is the trace vector and the derivatives are symmetrized over μ, ν, ρ.","section":"Sec. 5.3, Eq. (98)"},{"comment":"In the denominator of r_±^D, the expression \"2 - aD ∓ D√f_D\" is clearer with parentheses: (2 - aD ∓ D√f_D).","section":"Sec. 5.1, Eq. (70)"},{"comment":"The phrase \"we comment on that point later\" is a forward reference without a specific location; please refer explicitly to the paragraph in Sec. 5.3 where the restriction A ≠ 1 is justified.","section":"Sec. 5.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for a hep-th journal and the central idea is attractive. The main obstacle is the missing derivation of the spin-3 amplitude (22), which is the foundation of the spin-3 equivalence theorem. If the authors can supply that derivation (and the details for the special cases), the result would be convincing. I do not see an internal inconsistency in the algebraic core of Sec. 5, but the omitted computation is precisely the load-bearing step, so major revision is warranted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper has one genuinely new result: the generalized spin-3 TDiff model of section 3 and the theorem in section 5 that, within the ghost-free region f>0, every model in the two-parameter family is physically equivalent to the Maxwell-like model via invertible source redefinitions. The analog for spin-2 was partly known, but the spin-3 case, including the two-point amplitude (22) and the explicit r-shift/s-shift construction, is new. The Lagrangian diagonalizations (37), (47), (55) independently corroborate the particle content and the sign of f, which is good evidence.\n\nThe authors also work out the connection to the tensionless string doublet action carefully, and they are honest about the nonlocal redefinitions: they say plainly they have not been able to justify them rigorously. That is a real limitation, but it is acknowledged, not hidden.\n\nNow the soft spots. The main one is exactly what the stress-test note identifies: equation (22) is load-bearing and it is asserted rather than derived. The text says the projection-operator computation was carried out, the appendix gives the operator basis, but the actual contraction leading to the spin-1 residue with coefficient 3/(D f_{D+2}) and the source combination J^T = (2 - a(D+2))J + (a-1)T is not shown. Special cases a=1 and a=2/(D+2) are dismissed with 'basically the same expression' and 'we recover (22)'. Section 5.2 then uses exactly those coefficients to identify the ghost-free region and to derive B=A^2. If there is an error in that amplitude, the theorem fails. I have no reason to believe there is an error—the diagonalized Lagrangians point the right way—but a referee should require the full computation. Second, the nonlocal field redefinition of section 5.3 is verified with Mathematica, but no code or notebook is shipped, so the claim that it works for arbitrary s3 is not independently checkable from the text. That is minor.\n\nWho is this for? People working on TDiff modified gravity, higher-spin free models, and the tensionless string limit. It deserves a serious referee. The referee's main job is the projection-operator computation. I would accept it for review and, if the derivation comes back clean, cite it.","headline":"Solid, useful paper with a real equivalence theorem; the main proof rests on an amplitude (Eq. 22) that is stated but not derived—send to a referee who will demand that computation.","tokens_in":20750,"tokens_out":3235,"would_cite":true,"duration_ms":31864,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["11.25.-w","11.10.Ef"],"model":"deepseek-v4-flash","headline":"All ghost-free TDiff spin-2 and spin-3 models are physically equivalent.","keywords":["transverse diffeomorphisms","TDiff models","spin-2","spin-3","Maxwell-like models","tensionless string limit","two-point amplitude","ghost-free parameter space"],"falsifier":"Compute the full rank-3 propagator from (9) at $c=c_S$ in a fixed dimension (say $D=4$) for a numerical point with $f_{D+2}>0$ and compare the saturated amplitude with (22). If the coefficient of $|\\tilde J^T_\\mu|^2$ differs from $3/(D f_{D+2})$, or if an additional transverse-traceless pole appears, the equivalence (74) fails. Equivalently, check whether the two root formulas (70) and (89) ever produce non-invertible r-shifts in $D\\ge 3$ with $f>0$; a counterexample would break the universal step.","tokens_in":19685,"feed_emoji":"🔁","tokens_out":13561,"duration_ms":108100,"temperature":0.7,"pith_summary":"The paper studies the most general local, massless, symmetric-tensor theories of spin-2 and spin-3 that are invariant under transverse diffeomorphisms (TDiff), where the gauge parameters are constrained to be divergence-free. It claims that within the ghost-free region of the parameter space — specified by $f_D(a,b)>0$ for spin-2 and $f_{D+2}(a,b)>0$ for spin-3 — every model is physically equivalent to every other, and in particular to the simplest 'Maxwell-like' model that keeps only the first two kinetic terms. The proof works by a local field redefinition (an r-shift) plus a redefinition of the sources (an s-shift), and the same two-step move connects the entire ghost-free family to the 'doublet' action that comes from the tensionless limit of open bosonic string field theory. If correct, this collapses a large family of alternative gravity and higher-spin models into one physical content, explaining why their extra lower-spin particles do not create observable differences except through how they couple to sources.","feed_headline":"All ghost-free TDiff spin-2 and spin-3 models are one theory","feed_subtitle":"A two-step source redefinition shows every ghost-free model reproduces the simplest one—the Maxwell-like form tied to string theory.","key_machinery":"The load-bearing object is the saturated, gauge-invariant two-point amplitude $A_2(k)$, whose pole structure shows a massless spin-$s$ particle plus a lower-spin companion whose sign is set by $f_D$ (spin-2) or $f_{D+2}$ (spin-3). The equivalence is carried by two transformations: local r-shifts, $h_{\\mu\\nu} \\to h_{\\mu\\nu} + r \\eta_{\\mu\\nu} h$ and the rank-3 analogue, which preserve the TDiff form and multiply $f$ by $(1+rD)^2$ (or $(1+r(D+2))^2$), and s-shifts of the sources that remove the last parameter and can be interpreted as a nonlocal field redefinition of the form $h \\to h + s\\,(\\partial\\partial h)/\\Box$. Combining them, any point with $f>0$ is brought to $(0,0)$, the Maxwell-like model, defined as the simplest TDiff action that keeps only the first two kinetic terms.","core_discovery":"On the paper's own terms, the central result is an exact equality of gauge-invariant two-point amplitudes: $A_2[a,b,T,J]=A_2[0,0,T(r,s),J(r,s)]$ whenever $f_D(a,b)>0$, and the spin-3 analogue $A_2[a,b,c_S,T,J_\\mu]=A_2[0,0,0,T(r,s),J(r,s)]$ whenever $f_{D+2}(a,b)>0$. Here $f_D=(D-2)(a^2-b)+(a-1)^2$ and $f_{D+2}=D(a^2-b)+(a-1)^2$ control whether the companion spin-0 (spin-1) particle is physical or a ghost, and $c_S=3a^2-2b$ is fixed by TDiff invariance. The r-shift maps the parameters along a curve that scales $f$ by a positive factor; the s-shift redefines sources so that the remaining parameter dependence cancels. Consequently any ghost-free TDiff spin-2 or spin-3 model has the same free on-shell content as the Maxwell-like model, and the models differ only by a BRST-cohomologically trivial term.","pith_inferences":["The same r-shift/s-shift construction should generalize to symmetric rank-$s$ TDiff models, with the companion of spin $s-2$ and a function $f_{D+s-2}$; if so, Maxwell-like models are the unique ghost-free TDiff representatives for all spins, not just $s=2,3$.","Because the proof is on-shell (amplitudes), the equivalence may not persist off-shell for quantities such as energy-momentum tensors or locality; testing cubic vertices would decide whether the whole ghost-free family is one interacting theory.","The special point at $a=2/(D+2)$ is not reachable by local r-shifts, so the paper's equivalence relies on nonlocal redefinitions there; this suggests the ghost-free subspace is an amplitude-equivalence class rather than a gauge-equivalence class of actions."],"forward_implications":["For spin-2, every TDiff model with $f_D>0$ predicts exactly the same free amplitudes as the Maxwell-like model, so the scalar companion cannot distinguish models inside the ghost-free region; only the source couplings change.","For spin-3, every model with $f_{D+2}>0$ contains one healthy massless spin-3 particle and one healthy massless spin-1 particle, with no ghosts.","The apparent mismatch between generic TDiff models and the tensionless-string doublet action disappears: after the amplitude equivalence, all ghost-free TDiff models match the string-derived doublet spectrum.","At $f=0$ (or $f_{D+2}=0$) a larger gauge symmetry appears and the lower-spin companion decouples, so the boundary between physical and unphysical regions is also the boundary of the equivalence class."],"supporting_citations":[{"why":"Establishes the spin-2 TDiff symmetry, its on-shell amplitude, and the ghost-free condition used throughout.","marker":"[11]"},{"why":"Supplies the spin-2 two-point amplitude, the r-shift map, and the source constraint reused by the equivalence proof.","marker":"[17]"},{"why":"Defines the Maxwell-like TDiff models for arbitrary spin, the target models to which every ghost-free model is claimed equivalent.","marker":"[12]"},{"why":"Provides the doublet action from the tensionless limit of string theory that the TDiff models are matched against.","marker":"[16]"},{"why":"Gives the rank-3 projection-operator basis used to derive the spin-3 gauge-invariant amplitude (22).","marker":"[24]"},{"why":"Defines the Fronsdal spin-3 model used as the pure-spin-3 reference point.","marker":"[8]"},{"why":"Defines the transverse-invariant Skvortsov-Vasiliev model that fixes the special point a=2/(D+2).","marker":"[9]"}],"fun_headline_variants":["Ghost-free TDiff spin-2/3 models unify into Maxwell-like theory","All ghost-free TDiff spin-2 and spin-3 reduce to Maxwell-like","Ghost-free TDiff spin-2 and spin-3 are equivalent to Maxwell-like","Unified ghost-free TDiff spin-2 and spin-3 as Maxwell-like","Ghost-free TDiff spin-2/3: all equivalent to Maxwell-like"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The spin-3 gauge-invariant two-point amplitude (22), whose projection-operator derivation is only sketched, is correct and complete; if a coefficient is wrong or a spin-1 term is missing, the source-redefinition equivalence and the ghost-free condition $f_{D+2}>0$ would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Ghost-free TDiff spin-2/3 models unify into Maxwell-like theory","All ghost-free TDiff spin-2 and spin-3 reduce to Maxwell-like","Ghost-free TDiff spin-2 and spin-3 are equivalent to Maxwell-like","Unified ghost-free TDiff spin-2 and spin-3 as Maxwell-like","Ghost-free TDiff spin-2/3: all equivalent to Maxwell-like"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000714,"raw_usage":{"total_tokens":3246,"prompt_tokens":1013,"completion_tokens":2233,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":629,"completion_tokens_details":{"reasoning_tokens":2128}},"tokens_in":629,"tokens_out":2233,"duration_ms":14506,"temperature":1.0,"reasoning_tokens":2128,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T21:39:53.639420+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the full rank-3 propagator from (9) at $c=c_S$ in a fixed dimension (say $D=4$) for a numerical point with $f_{D+2}>0$ and compare the saturated amplitude with (22). If the coefficient of $|\\tilde J^T_\\mu|^2$ differs from $3/(D f_{D+2})$, or if an additional transverse-traceless pole appears, the equivalence (74) fails. Equivalently, check whether the two root formulas (70) and (89) ever produce non-invertible r-shifts in $D\\ge 3$ with $f>0$; a counterexample would break the universal step.","supporting_citations":[{"cited_title":"Transverse Invariant Higher Spin Fields,","cited_arxiv_id":null,"evidence_quote":"Defines the transverse-invariant Skvortsov-Vasiliev model that fixes the special point a=2/(D+2)."}],"review_version":1}