{"id":"5e31db9a-3bee-492c-99bc-48fca52e197b","arxiv_id":"2412.05575","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A recursive framework built on the optical theorem and a doubled-field action computes differential cross-sections directly, reproducing known tree-level 2 to 2 and 2 to 4 phi^4 results.","lead":"The paper builds a new way to compute particle collision rates that skips the usual step of squaring quantum amplitudes, instead calculating the rates directly from loop diagrams using a recursive technique. If it works in realistic theories like QCD, it could make high-multiplicity scattering predictions cheaper to compute, but the paper only demonstrates it on simple scalar toy models.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The completeness of the Section 3.4 filtering algorithm is unproven and load-bearing for the claimed exact match of the 2→4 cross-section.","rationale":"The paper presents a coherent framework and reproduces two known cross-sections, which is genuine evidence that the recursion and filtering are at least self-consistent for those cases. The optical theorem connection, Eq. (3.1), is standard, and the cut-diagram counting conditions (3.38)-(3.40) are natural. Nonetheless, the central claim 'we confirmed that these two quantities match exactly' is only as strong as the completeness of the filtering step. The algorithm is illustrated on two representative diagrams but not proven for the full set of topologies appearing at L0=3, n=4; the paper does not enumerate all three-loop four-cut diagrams, nor does it show the algebraic reduction from Eq. (4.4) to Eq. (4.7). These are addressable, concrete gaps, and an independent enumeration or symbolic reduction would settle them. Therefore the reader's CONDITIONAL verdict is appropriate: the framework is plausible and partially validated, but the load-bearing completeness assumption is not yet established.","tokens_in":20965,"tokens_out":15432,"duration_ms":150072,"concrete_test":"Independently generate all cut diagrams contributing to A(3)1+2-3-4+ by applying the LTE cutting rules to the three-loop off-shell current, then apply the Section 3.4 filter and compare the surviving integrand term-by-term with Eq. (4.4). A mismatch in either direction falsifies the completeness claim. Alternatively, symbolically reduce Eq. (4.4) to the factorized form in Eq. (4.7); if the reduction does not produce the product of the two 10-term tree sums with symmetry factor 1/4!, the claimed exact match fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central validation is the claimed exact match of the one-loop and three-loop doubled amplitudes with dσ(0)2→2 and dσ(0)2→4. This rests on the filtering algorithm in Section 3.4: after steps (1)-(4), the paper states that the remaining terms contribute to σ(L)2→n only (lines following Eq. (3.40)). The filter uses the cut-count condition N+- + N-+ = n (3.38), the rank condition L = L0 - rank M (3.40), and the connectedness condition C = 2, with C given by (3.39). These conditions are demonstrated on exactly two diagram types, (a) and (b) of Figure 3, and no proof is given that they are complete for arbitrary loop order and multiplicity. If the filter misses terms or admits contaminants, the nine terms in Eq. (4.4) would not be the full A(3) integrand, and the asserted equality with the known dσ(0)2→4 in Eq. (4.7) would be unsupported. The paper also states the match is 'confirmed exactly' without showing the reduction of (4.4) to (4.7), so the completeness of the filter is the key unverified step on which the central claim depends.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript presents a recursion-based framework for computing differential cross-sections directly from loop-level amplitudes, bypassing the standard step of squaring scattering amplitudes. The construction doubles the field content of φ⁴ theory according to the Schwinger–Keldysh prescription, derives Dyson–Schwinger equations in the doubled theory, and builds quantum off-shell recursion relations up to three loops (Section 3). The doubled amplitudes A^{(1)} and A^{(3)} obtained from these recursions are claimed to reproduce exactly the tree-level differential cross-sections dσ^{(0)}_{2→2} and dσ^{(0)}_{2→4} (Section 4.2, Eqs. (4.5) and (4.7)). A filtering algorithm (Section 3.4) is introduced to select the terms in the doubled amplitude that belong to a chosen σ^{(L)}_{2→n}. The central validation rests on two unshown steps: the completeness of the filtering algorithm, demonstrated only on two diagram topologies, and the reduction of the nine-term integrand (4.4) to the squared-amplitude form (4.7), which is asserted rather than presented.","tokens_in":21303,"tokens_out":37547,"duration_ms":342969,"significance":"If the central claim holds, the paper offers a proof of principle for a genuinely different computational route to cross-sections: the optical theorem identifies the cross-section with a loop amplitude, and the quantum off-shell recursion computes that amplitude algebraically without amplitude squaring, with potential gains for the color/helicity sums of gauge theories. The paper's concrete deliverables—the doubled-action DS equations, the recursion relations up to three loops, the filtering conditions (3.38)–(3.40), and the explicit nine-term integrand (4.4)—are checkable, and the one-loop equality (4.2)→(4.5) is simple enough to verify by hand and is consistent with the known tree amplitude. Credit is due for carrying the recursion to three loops and for illustrating the filter on two explicit diagram types. However, the manuscript is purely analytic: there are no machine-checked algebraic reductions, code, or numerical checks, and the two load-bearing steps (filter completeness; the reduction (4.4)→(4.7)) are asserted, not shown.","major_comments":[{"comment":"The completeness of the filtering algorithm is load-bearing and unproven. Steps (1)–(4) are validated on exactly two diagram topologies, (a) and (b) of Figure 3, and the text then asserts that “the remaining terms contribute to σ^{(L)}_{2→n} only.” No argument establishes that condition (3.38) (N^{+-}+N^{-+}=n), the rank condition (3.40) (L=L0−rank M), and the connectedness condition C=2 from (3.39) are necessary and sufficient for every topology at arbitrary loop order and multiplicity. Concretely: the rank of M is never analyzed beyond the two examples, so degenerate configurations (rank M < n−1) are not discussed; (3.38) fixes only the total number of cuts, and the paper never shows whether the recursion generates terms with N^{+-}=2 or 3 that would also pass the filter (all nine terms in (4.4) have N^{+-}=1); and step (4) of the algorithm is logically redundant with step (3), since (3.39) defines C. If the filter misses target terms or admits contaminants, the “nine relevant terms” of (4.4) would not be the full A^{(3)} integrand and the asserted equality with dσ^{(0)}_{2→4} would be unsupported.","section":"Section 3.4"},{"comment":"The central validation—“we calculated the doubled amplitude ... and compared it with the differential cross-section ... We confirmed that these two quantities match exactly”—is asserted, not demonstrated. The text does not show the reduction of (4.2) to (4.5) nor of the nine-term integrand (4.4) to the squared-amplitude form (4.7). In particular, it does not specify which products of the two ten-term propagator sums appearing in (4.7) each of the nine terms of (4.4) produces after the on-shell integrations and momentum relabelings, nor how the symmetric factors 1/2 and 1/4! combine with the relative coefficients (1/2, 1/6, 1/4, 1) in (4.4). Equation (4.7) also switches from “∼” to “=” without comment. Because the exact match with known squared amplitudes is the only evidence that the recursion-plus-filter pipeline computes the correct cross-section, the authors should include this reduction (an appendix tabulating the mapping of the nine terms onto the products in (4.7)) or provide a symbolic/numerical verification at generic external kinematics.","section":"Section 4.2, Eqs. (4.4)–(4.7)"},{"comment":"The paper does not show how the quoted results (4.2) and (4.4) follow from the recursion relations (3.21)–(3.35). The derivation of (4.2) from (3.25)–(3.26) is straightforward and should be given as an illustration of the pipeline. For the three-loop case, Section 3.3.4 states “we skip explicitly listing the ℏ³-order terms” and Section 4.1 states that the filtered current “has only nine relevant terms” without presenting the intermediate recursion output, the total number of generated terms before filtering, or any code or ancillary material. Since the nine-term list (4.4) is the empirical basis of the paper's central claim, the reader currently cannot verify that it is the complete filtered output of the recursion. Please provide the intermediate output or a reproducible computation.","section":"Sections 3.3–4.1"},{"comment":"The object compared with the doubled amplitude is not the standard differential cross-section. The standard dσ for 2→n carries the flux factor 1/(2E_A 2E_B |v_A−v_B|), which appears in the paper's own summary of the optical theorem, Eq. (A.6). The right-hand sides of (4.5) and (4.7) are dimensionless phase-space integrals of |M|² (with symmetric factors), i.e., precisely the right-hand side of the optical theorem (A.4), whereas a cross-section has mass dimension −2 and contains the flux factor. The paper should either include the flux factor, or explicitly state that it computes the reduced cross-section and specify the normalization of M^{(0)} used in the comparison. As written, the claim that the method “reproduces the differential cross-section” is a convention-dependent statement that the reader cannot check.","section":"Section 4.2 and Eq. (A.6)"}],"minor_comments":[{"comment":"In the paragraph before Eq. (4.3), “Φ^{(2)}_{1+2−3−}” should be a three-loop object (Φ^{(3)} or A^{(3)}); the two-loop label contradicts the surrounding three-loop discussion and the statement A^{(2)} ∼ 0 in the same paragraph.","section":"Section 4.1"},{"comment":"The matrix M whose rank defines l is introduced by example only; please state the general construction (rows are the coefficient vectors of the cut-line momenta in the basis of loop momenta), state explicitly that l is the dimension of the span of the cut momenta, and explain how degenerate configurations are treated.","section":"Section 3.4, Eqs. (3.40)–(3.42)"},{"comment":"The notation for momentum-space propagators switches from “˜D” in (4.4) to “D” in (4.7) without comment; please use a single convention.","section":"Section 4.2"},{"comment":"The symbol “∼”, defined in (4.3) as equality after discarding terms that do not contribute, is used in the first line of (4.7) where an exact equality with the squaring result is claimed; please use “=” consistently throughout.","section":"Section 4.2, Eq. (4.7)"},{"comment":"Please state explicitly that (3.1) is the optical theorem taken as input from S-matrix unitarity, so the paper's contribution is the recursive construction of the doubled amplitude (the left-hand side), not an independent derivation of the theorem; this would clarify the scope of the validation.","section":"Section 3.1, Eq. (3.1)"},{"comment":"The truncation of the DS hierarchy at fixed ℏ order (“higher descendant fields always carry higher ℏ orders”) is asserted without proof; a brief justification or a pointer to the corresponding argument in Ref. [4] would be helpful.","section":"Section 2.3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a v1 with a clear proof-of-principle structure and no obvious fatal error; my concerns are about missing demonstrations rather than wrong results. I would ask the authors, in revision, to (i) prove or substantially extend the validation of the filtering algorithm, (ii) show the reduction (4.4)→(4.7) or provide a machine-checkable or numerical verification, and (iii) fix the cross-section normalization convention. If these are provided, the paper is publishable as a techniques paper in JHEP; the topic fits the journal's scope, given the recent QOR literature (Refs. [4, 5]). One editorial note: the abstract's claim “We confirmed that these two quantities match exactly” should be calibrated until the supporting computation is actually shown."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nHere's my read of arXiv:2412.05575. The paper is worth taking seriously, but the exact-match claim is only as strong as a filtering algorithm whose completeness is never proven.\n\nWhat's genuinely new: combining the Schwinger-Keldysh doubling prescription with quantum off-shell recursion to generate largest-time-equation cut diagrams, then filtering them by loop order to get differential cross-sections directly. That combination is not in the cited literature, and the two examples are real checks: a one-loop doubled amplitude reproducing the 2→2 tree cross-section, and a three-loop calculation producing nine cut diagrams claimed to reproduce the 2→4 cross-section. If correct, this is a legitimate alternative to squaring amplitudes and, in principle, avoids color/helicity sums. The recursion machinery is derived cleanly from the DS equations, and the perturbiner expansion is a sensible generating device.\n\nThe soft spots are real. The filtering algorithm in Sec. 3.4 is load-bearing: after steps (1)-(4), the paper states that 'the remaining terms contribute to σ(L)2→n only,' but the conditions (3.38)-(3.40) are only checked on two diagram types, (a) and (b) of Fig. 3. No argument is given that they are sufficient for all loop orders and multiplicities. If the filter admits contaminants or drops terms, the nine terms in (4.4) are not the full integrand, and the equality (4.7) would be unsupported. That matters because the paper asserts the match is 'exact' without showing the reduction from (4.4) to (4.7). Likewise, the one-loop result (4.2) is stated with no derivation. For a methods paper, that algebra should be in an appendix or supplementary file.\n\nTwo smaller issues: the claimed efficiency for QCD and the Standard Model is speculative, with no color or helicity example attempted; and there is a typo in Sec. 4.1 where 'Φ(2)' should be 'Φ(3)', otherwise the sentence contradicts the preceding paragraph.\n\nThe citation pattern is fine; the self-citations are to Lee's earlier recursion work, which this directly builds on. The reader's circularity concern is a non-issue: the construction uses unitarity, it does not pretend to derive it.\n\nBottom line: this is a promising methods paper with two non-trivial validations. The missing proof of filter completeness and the unshown algebra are addressable, not fatal. If I were the editor, I'd send it to a referee with a request for a proof or more systematic check of the filter and an explicit demonstration of the key equalities. It deserves a serious referee, but not a rubber stamp.\n\nI'd bring it to a reading group only if people are working on recursive amplitude methods.","headline":"A genuinely new recursion-based route to differential cross-sections, with solid two-example validation but an unproven filtering step that the exact-match claim rests on.","tokens_in":21731,"tokens_out":3450,"would_cite":true,"duration_ms":32159,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that a differential cross-section can be computed directly as a loop amplitude via the optical theorem, bypassing the usual step of squaring scattering amplitudes and summing over colors and helicities.","keywords":["optical theorem","differential cross-section","largest time equation","quantum off-shell recursion","quantum equations of motion","off-shell currents","phi-four scalar theory","cut diagrams"],"falsifier":"Compute the two-loop doubled amplitude $A^{(2)}_{1+2-3-4+}$ in $\\phi^4$ and apply the filter targeting $\\sigma^{(1)}_{2\\to2}$; the surviving integrand should equal $2\\,\\mathrm{Re}\\bigl[(M^{(0)}_4)^* M^{(1)}_4\\bigr]$. If any term survives that is not in that one-loop cross-section, or any required term is missing, the filter's completeness claim is false.","tokens_in":1713,"feed_emoji":"⚛️","tokens_out":2588,"duration_ms":108689,"temperature":0.7,"pith_summary":"The paper presents a new way to compute differential cross-sections in quantum field theory without first computing and then squaring scattering amplitudes. Instead, it uses the optical theorem to rewrite a cross-section as a loop amplitude in a theory with doubled fields. The authors develop a recursive scheme that generates this doubled amplitude directly from quantum equations of motion, and they validate it by exactly reproducing the tree-level $2\\to2$ and $2\\to4$ differential cross-sections in $\\phi^4$ theory. If correct, this turns cross-section calculations into recursion problems that naturally handle color and helicity sums, offering a potentially efficient alternative for high-multiplicity processes.","feed_headline":"Cross-sections from loop amplitudes, no squaring needed","feed_subtitle":"Doubled-field recursion plus the optical theorem reproduces tree-level 2-to-2 and 2-to-4 scattering exactly.","key_machinery":"The load-bearing object is the doubled amplitude $A_{1+2-3-4+}$, a $2\\to2$ amplitude in a field theory containing two copies of the scalar field, $\\phi^+$ and $\\phi^-$, with doubled propagators that include on-shell pieces. The identity that carries the argument is $A_{1+2-3-4+}=\\sum_X M(1+4+\\to X)(M(2-3-\\to X))^*$, the optical theorem written at amplitude level: the right-hand side is exactly the squared matrix element summed over intermediate states that defines a differential cross-section. The mechanism is a recursive construction: the doubled action gives equations of motion for correlation functions, a perturbiner expansion (a plane-wave ansatz generating off-shell currents for each multi-particle word) turns those equations into off-shell recursion relations, and the largest time equation, an algebraic form of the optical theorem at diagram level, identifies which components carry cuts. A filtering algorithm then keeps only the integrands with the number of on-shell cuts $N^{+-}+N^{-+}=n$, two connected sectors $C=2$, and loop count $L=L_0-\\operatorname{rank} M$, which are the terms contributing to the desired $\\sigma^{(L)}_{2\\to n}$.","core_discovery":"The discovery is that the differential cross-section for a $2\\to n$ process is not something to be obtained by squaring the tree amplitude; it is the on-shell content of a $2\\to2$ amplitude in a theory with doubled fields, evaluated at $(n-1)$-loop order. The doubled amplitude $A^{(L)}_{1+2-3-4+}$ satisfies the optical-theorem relation $A_{1+2-3-4+}=\\sum_X M(1+4+\\to X)(M(2-3-\\to X))^*$, and the paper shows how to generate this object algebraically by solving the quantum equations of motion of the doubled action with quantum off-shell recursions. The largest time equation supplies the cutting rule: propagators of type $\\Delta^{\\pm}$ put internal lines on shell, so the computed amplitude automatically contains the phase-space integrals that define the cross-section. The paper validates the scheme in $\\phi^{4}$ theory by exact matching: the one-loop result equals the tree $2\\to2$ differential cross-section and the three-loop result equals the tree $2\\to4$ differential cross-section, after a filtering algorithm removes contributions from other cross-sections at the same loop order.","pith_inferences":["If the filtering algorithm is complete to all orders, the method should reproduce loop-level cross-sections as well, for example the one-loop $2\\to2$ cross-section from the two-loop doubled amplitude, which the paper notes but does not compute.","The rank condition $L=L_0-\\operatorname{rank} M$ on the loop-momentum matrix is likely the seed of a general algebraic cut-selection rule that could be applied to fermions and gauge fields without inventing new diagram topologies.","The recursion may also give a natural way to combine cross-section computation with Monte Carlo phase-space integration, since it already produces the cut integrand in a factorized form.","Connecting the doubled action with real-time thermal or nonequilibrium correlation functions could let the same machinery compute scattering in media, although the paper does not explore that direction."],"forward_implications":["Cross-section calculations stop needing the explicit $O(N^2)$ squaring of $N$-term amplitudes; the recursion produces the squared object directly.","Color and helicity sums are folded into the doubled-field recursion instead of being performed separately, which could make gauge-theory cross-sections more tractable.","The same recursion scheme works for any theory with a local action, so extending the doubled action to QCD or the Standard Model is a direct next step.","The filtering conditions give an algebraic, rather than diagram-by-diagram, way to decide which loop orders contribute to a given $2\\to n$ tree-level cross-section.","Because the output is the cross-section integrand with phase-space delta functions, a further phase-space integration yields total cross-sections."],"supporting_citations":[{"why":"states the optical theorem used for the key identity.","marker":"[1]"},{"why":"introduces the quantum off-shell recursion method the paper builds on.","marker":"[4]"},{"why":"extends the off-shell recursion to higher-loop currents and gives the bracket notation.","marker":"[5]"},{"why":"introduces the off-shell current recursion at tree level that this work generalizes.","marker":"[6]"},{"why":"gives the diagrammatic cutting rules that the largest time equation replaces.","marker":"[12]"},{"why":"formulates the largest time equation used to encode cuts.","marker":"[13]"},{"why":"lays out the doubled-action formalism that the doubling prescription adopts.","marker":"[19]"}],"fun_headline_variants":["Loop recursion yields cross-sections, no squaring","Optical theorem + recursion = cross-sections","Cross-sections directly from loop amplitudes","No amplitude squaring: loop recursion does it"],"cache_read_input_tokens":23936,"weakest_assumption_plain":"The filtering algorithm is assumed to be complete: it has been checked on two diagram types only, so the paper has not proven that, for any loop order and final-state multiplicity, its three conditions keep exactly the target terms and nothing else.","fun_headline_variants_meta":{"raw":{"variants":["Loop recursion yields cross-sections, no squaring","Optical theorem + recursion = cross-sections","Cross-sections directly from loop amplitudes","No amplitude squaring: loop recursion does it"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000417,"raw_usage":{"total_tokens":2164,"prompt_tokens":975,"completion_tokens":1189,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":591,"completion_tokens_details":{"reasoning_tokens":1131}},"tokens_in":591,"tokens_out":1189,"duration_ms":8624,"temperature":1.0,"reasoning_tokens":1131,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T20:34:58.979440+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the two-loop doubled amplitude $A^{(2)}_{1+2-3-4+}$ in $\\phi^4$ and apply the filter targeting $\\sigma^{(1)}_{2\\to2}$; the surviving integrand should equal $2\\,\\mathrm{Re}\\bigl[(M^{(0)}_4)^* M^{(1)}_4\\bigr]$. If any term survives that is not in that one-loop cross-section, or any required term is missing, the filter's completeness claim is false.","supporting_citations":[{"cited_title":"Feenberg, The Scattering of Slow Electrons by Neutral Atoms , Phys","cited_arxiv_id":null,"evidence_quote":"states the optical theorem used for the key identity."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the diagrammatic cutting rules that the largest time equation replaces."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"formulates the largest time equation used to encode cuts."},{"cited_title":"chao Chou, Z","cited_arxiv_id":null,"evidence_quote":"lays out the doubled-action formalism that the doubling prescription adopts."}],"review_version":1}