{"id":"07d2061c-c02e-486d-8cd9-89147bb47bd2","arxiv_id":"1908.06357","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"Forward-proton-tagged photon-induced production of a leptophobic Z' mediator would exclude mZ' below about 1.4 TeV at 14 TeV and 3000 fb^-1, with a dark matter mass bound near 550 GeV for the vector benchmark.","lead":"This paper calculates how well the LHC could see dark matter produced together with a jet when one of the colliding protons stays intact and is caught by forward detectors. It finds this photon-induced channel could rule out a leptophobic Z' mediator up to roughly 1.4 TeV, if pileup backgrounds can be suppressed.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Projected reach is conditional on an untested HL-LHC pileup-suppression mechanism; Section 3 explicitly defers it to future work.","rationale":"The reader's weakest assumption correctly identifies the most load-bearing condition of the central claim. The paper itself flags the pileup problem explicitly in Section 3 and leaves the suppression mechanism to future work, so this is not a manufactured concern. The statistical procedure otherwise appears internally consistent: the event counts are reproduced from public tools, the xi cut is motivated by the distributions, and the projected contours follow from NS/sqrt(NB) > 1.96. The paper's own final paragraph also notes that existing dijet searches exclude mZ' up to about 5 TeV, which is much stronger than the projected 1.4 TeV reach; this further limits the scientific novelty and reinforces that the work should be read as a conditional feasibility study. However, the internal calculation is not invalidated by that external constraint, and the conditional verdict already captures the main risk. Since the reader's verdict is CONDITIONAL and the identified pileup assumption is the same one that would need to be resolved, no verdict change is warranted.","tokens_in":21145,"tokens_out":13845,"duration_ms":156715,"concrete_test":"Run a full HL-LHC simulation with PYTHIA pileup overlay at mu = 50 and mu = 200, 14 TeV, and 3000 fb^-1, through AFP/CT-PPS forward-proton reconstruction with the proposed 0.05 < xi < 0.15 selection and timing/vertex-matching cuts. Count j + ETmiss events with an accidental forward proton surviving all cuts. If this count is not suppressed to the level of the ~10 expected photon-induced background events, the projected exclusion reach in Fig. 5 fails. If a full simulation is not available, compute the required rejection factor from the inclusive pp -> j + ETmiss cross section (~12 pb) and compare with published timing-based forward-proton tag rejection capabilities; a rejection factor below ~10^3 would falsify the assumed background control.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central sensitivity numbers (mZ' < 1.4 TeV and mchi > 550 GeV at mZ' = 1.2 TeV, vector scenario) are obtained using only the irreducible photon-induced background pp -> p gamma p -> p j nu nubar X. Section 3 states that inclusive pp -> j + ETmiss X with an accidental forward proton from pileup is about 1000 times larger than the photon-induced signal and that 'unless the pileup events are controlled well enough, the process pp to j/ET X with one final forward proton from pileup events overwhelmingly dominates over our process.' The suppression mechanism is not demonstrated; the paper says 'we assume that the pileup events are sufficiently suppressed, and we leave an investigation of the suppression mechanism to the future work.' Without a quantitative estimate of the accidental forward-proton tag rate under HL-LHC conditions (mu = 50-200, AFP/CT-PPS timing and vertex matching), the exclusion contours in Fig. 5 are not guaranteed to be observable. The abstract presents the sensitivity without this condition. This is a missing support rather than an internal error, but it is load-bearing: if the required pileup rejection cannot be achieved, the channel does not produce the assumed signal-to-background ratio.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a parton-level feasibility study of fermionic dark-matter pair production through photon-induced interactions at the LHC, pp -> p gamma p -> p j chi chi-bar X, with a leptophobic spin-1 mediator Z' and one intact proton tagged by forward proton detectors such as AFP and CT-PPS. Signal and irreducible Standard Model background (pp -> p gamma p -> p j nu nu-bar X) are simulated with MadGraph5_aMC@NLO using the DMsimp model and NNPDF2.3, with the selections pT(j) > 200 GeV, |eta(j)| < 3.0, and 0.05 < xi < 0.15. Exclusion limits are derived with the criterion N_S / sqrt(N_B) > 1.96 at 95% C.L. for sqrt(s) = 14 TeV and L_int = 3000 fb^-1, and the paper claims sensitivity to m_Z' up to about 1.4 TeV and, in the vector scenario, a lower bound m_chi > about 550 GeV at m_Z' = 1.2 TeV. The analysis explicitly omits pileup, single-diffractive, and other QCD backgrounds, and the pileup suppression needed for the central claim is left to future work.","tokens_in":21438,"tokens_out":9120,"duration_ms":98422,"significance":"If the projected sensitivity can be made realistic, the forward-proton-tagged photon-induced channel would be a useful complement to monojet and dijet searches, potentially covering a mixed vector-axial mediator scenario that the energetic-jet analyses have not presented. The paper has clear strengths: the benchmark couplings follow LHC Dark Matter Working Group recommendations, the cut flow in Table 1 is explicit, the comparison with the ATLAS energetic-jet limits in Fig. 5 is informative, and the parton-level computation is internally consistent. The caveat is that the headline exclusion contours are computed with only the irreducible background and therefore depend on an unquantified assumption that pileup-generated forward protons can be rejected. Because that assumption is load-bearing, the published claims need either a quantitative treatment of the pileup background or a prominently qualified statement of what is being assumed.","major_comments":[{"comment":"This is the central issue. The paper states that pp -> j + /ET + X with an accidental forward proton from pileup has a rate about 1000 times larger than the tagged photon-induced process and that 'unless the pileup events are controlled well enough, the process pp -> j/ET X with one final forward proton from pileup events overwhelmingly dominates over our process.' It then explicitly says that the suppression mechanism is left to future work. Since N_B in the significance calculation includes only the irreducible photon-induced background, the 95% C.L. contours in Fig. 5, and the abstract's m_Z' < about 1.4 TeV and m_chi > about 550 GeV claims, rely on an unquantified assumption of essentially perfect rejection of accidental forward protons. The authors should include a quantitative estimate of the accidental tagging rate for HL-LHC pileup conditions (with timing and vertex-matching information for AFP and CT-PPS), add this component to the background in the N_S/sqrt(N_B) calculation, or present all sensitivity numbers as explicitly conditional on a stated pileup-rejection performance. A corresponding qualifier in the abstract is also needed.","section":"Section 3, pileup paragraph; Section 4 and Fig. 5"},{"comment":"The paper acknowledges that the combined dijet invariant-mass search excludes m_Z' above about 5 TeV for the benchmark quark couplings, which is much stronger than the forward-proton channel's claimed sensitivity of m_Z' < about 1.4 TeV. Because the exclusion contours in Fig. 5 lie almost entirely in parameter space already excluded by dijet searches, the statement that this production channel is 'sensitive' to the simplified model is potentially misleading. The authors should either identify a parameter region or observable that is not already covered by dijet bounds (for example, the DM mass reach at a fixed, still-allowed m_Z', or a smaller g_q scenario that evades dijet limits), or present the paper explicitly as a detector-feasibility and method study rather than as a new competitive constraint. This is a framing issue, but it directly affects the significance of the central claim.","section":"Section 4, dijet comparison paragraph"}],"minor_comments":[{"comment":"The abstract should carry the same caveat as Section 3 about the pileup background, since the present wording presents the sensitivity numbers as unconditional.","section":"Abstract and Section 3"},{"comment":"The sentence 'this cross section is about 2000 times smaller' should specify that it is about 2000 times smaller than the leading background cross section, to avoid ambiguity.","section":"Section 3, background sentence"},{"comment":"The axis labels in the figure captions are garbled (for example, 'Entries/GeV -1: 3000 fbint=14 TeV, Ls...') and should be cleaned up; the y-axis unit also appears to be 'Entries' in Fig. 4 but 'Entries/GeV' in Fig. 3.","section":"Figures 3 and 4 captions"},{"comment":"References [40] and [88] are left as placeholders with 'Non-standard form, no INSPIRE lookup performed' and must be replaced with full bibliographic entries.","section":"References [40] and [88]"},{"comment":"The treatment of single-diffractive and QCD processes is only a one-sentence assertion; a quantitative statement of why these backgrounds are negligible in the chosen 0.05 < xi < 0.15 window would strengthen the background model.","section":"Section 3, background discussion"}],"recommendation":"major_revision","confidential_remarks":"The parton-level analysis itself is coherent and the numerical implementation appears reproducible. The decisive condition is the explicit pileup assumption: without a quantitative estimate of accidental forward-proton tagging, the projected limits are not yet supported as HL-LHC projections. This is a fixable missing support rather than an internal contradiction, so I recommend major revision rather than rejection. The dijet-exclusion context should also be addressed, as it bears on the significance of the claimed sensitivity."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a clean, honest parton-level feasibility study of DM pair production in photon-induced pp collisions with one tagged forward proton. The genuinely new piece is extending the earlier EFT treatment of the same final state (Ref. [65]) to the simplified leptophobic Z' model, including off-shell contributions, and deriving new 95% exclusion contours for the vector, axial-vector, and mixed benchmark scenarios. The mixed scenario is new relative to the ATLAS mono-jet analysis. The calculation itself is transparent and reproducible: standard MadGraph setup with NNPDF2.3 and the DMsimp model file, a clear cut flow, and a simple NS/sqrt(NB) limit. The paper also does something rare: it states the biggest caveat directly in Section 3, noting that inclusive pp -> j + ETmiss with an accidental forward proton from pileup is about 1000 times more abundant than the photon-induced signal and that the suppression mechanism is left to future work. So the stress-test note is accurate. The caveat is load-bearing, not cosmetic. The abstract presents the sensitivity numbers without this condition, which is the main thing to fix in revision. A referee should ask for a quantitative estimate of the accidental-tag rate at HL-LHC pileup, or at least a clear conditional statement in the abstract. The analysis also quotes only statistical errors; systematic uncertainties on S2, PDFs, and the xi window are not estimated, though for a parton-level projection that is less serious. There is a second soft spot that the authors do acknowledge: for the benchmark couplings used, dijet searches already exclude mZ' up to about 5 TeV, far beyond this channel's reach, so the 'constraints' on the vector scenario are not physically current. The paper is better read as a demonstration of the channel's potential than as a realistic limit. The mixed scenario result is the one piece without a direct dijet equivalent. All that said, the paper deserves a serious referee. It is a legitimate feasibility study, the literature context is well covered, and the main flaw is a missing support rather than an internal inconsistency. With a revised abstract and a quantitative pileup paragraph, it could be a useful reference for forward-proton DM searches. I would not cite it as a definitive exclusion, but I'd keep it on the radar.","headline":"A clean parton-level feasibility study whose headline reach is conditional on an untested pileup-suppression assumption; the mixed-scenario curves are new, but the phenomenological window is already closed by dijet searches.","tokens_in":21952,"tokens_out":3262,"would_cite":false,"duration_ms":34223,"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":"Forward-tagged photon collisions at the LHC could exclude a vector dark-matter mediator up to about 1.4 TeV, and a dark-matter mass above about 550 GeV.","keywords":["dark matter","forward proton tagging","simplified dark matter models","leptophobic vector mediator","photon-induced production","missing transverse energy","LHC","Z' mediator"],"falsifier":"A measurement of the rate of jet-plus-missing-energy events that carry a forward proton from pileup, under Run-III conditions with the same $0.05<\\xi<0.15$ cut, would settle the claim: if that rate exceeds the 2305 background events assumed in the 95% confidence calculation, the projected 1.4 TeV mediator exclusion does not survive, and if it can be suppressed below that level, the channel is viable.","tokens_in":20954,"feed_emoji":"⚛️","tokens_out":13922,"duration_ms":128012,"temperature":0.7,"pith_summary":"This paper argues that dark matter produced in photon-induced proton collisions, with one intact proton recorded in forward detectors, gives a new LHC search channel for simplified dark matter models with a leptophobic spin-1 mediator. At 14 TeV and 3000 inverse femtobarns, it projects 95% confidence exclusions of mediator masses up to about 1.4 TeV and a dark-matter mass bound of about 550 GeV for a 1.2 TeV mediator in the vector-coupling scenario. The channel's power comes from tagging the intact proton and cutting on its momentum loss, which suppresses the dominant jet-plus-neutrino background. If the projections hold, they add a photon-initiated, detector-tagged route to constraining the dark-matter sector at the LHC.","feed_headline":"Forward-tagged LHC events could probe dark mediator to 1.4 TeV","feed_subtitle":"Using one intact proton to tag photon collisions, the paper projects dark-matter limits at 3000 inverse femtobarns.","key_machinery":"The carrying mechanism is forward proton tagging through the momentum fraction loss $\\xi$, the fractional momentum carried away from the intact proton, with detector acceptance $0.015<\\xi<0.15$ and an analysis cut $0.05<\\xi<0.15$. The signal's $\\xi$ distribution shifts to larger values as $m_{Z'}$ grows, while the dominant background $pp\\to p\\gamma p\\to pj\\nu\\bar{\\nu}X$ peaks at low $\\xi$, so the cut suppresses that background to about 28% of its original rate while keeping about 48% of the benchmark signal. The photon flux is modeled by the equivalent photon approximation with survival probability $S^2=0.7$, and the mediator interactions are implemented with benchmark spin-1 couplings.","core_discovery":"The paper studies the simplified model in which a leptophobic spin-1 mediator $Z'$ couples to quarks and to a Dirac fermion dark matter candidate $\\chi$, and considers the photon-induced process $pp\\to p\\gamma p\\to pj\\chi\\bar{\\chi}X$, where the photon-emitting proton stays intact and is tagged by a forward proton detector while the central detector sees a jet and missing transverse energy. The irreducible background is $pp\\to p\\gamma p\\to pj\\nu\\bar{\\nu}X$ with neutrinos summed over three flavors. Using parton-level event generation, jet cuts $p_T^j>200$ GeV and $|\\eta^j|<3.0$, the forward-proton cut $0.05<\\xi<0.15$, and a 95% confidence requirement $N_S/\\sqrt{N_B}>1.96$ at 14 TeV with 3000 fb$^{-1}$, the paper claims that mediator masses below about 1.4 TeV are excluded for all three coupling scenarios, with lower dark-matter mass bounds of about 550 GeV at $m_{Z'}=1.2$ TeV in the vector scenario, 400 GeV at 1.1 TeV in the axial-vector scenario, and 500 GeV at 1.1 TeV in the mixed scenario.","pith_inferences":["The same momentum-loss tagging logic could be applied to other invisible or semi-invisible new-physics channels, since the background suppression is kinematic rather than model-specific.","A pileup-aware detector-level study using timing or vertex matching is the natural next step: it would convert these parton-level projections into a search strategy, or show that the assumed suppression is not achievable.","Given the strong dijet limits on the mediator, the practical window for this channel is likely the lower-mediator-mass, higher-dark-matter-mass corner where resonance searches are weaker; the paper notes the dijet bound but does not quantify the overlap."],"forward_implications":["A null result in the forward-tagged channel at 14 TeV with 3000 inverse femtobarns excludes mediator masses below about 1.4 TeV at 95% confidence in all three coupling scenarios when the dark matter is light.","In the vector scenario the channel sets a dark-matter mass bound of about 550 GeV at a 1.2 TeV mediator, slightly stronger than the compared energetic-jet analysis for mediator masses below about 1.2 TeV.","The mixed vector-plus-axial scenario, not covered by the cited energetic-jet searches, receives a projected dark-matter mass bound of about 500 GeV at a 1.1 TeV mediator.","Cutting the intact proton's momentum loss to the 0.05-0.15 window reduces the main background to about 28% of its original rate while keeping about 48% of the benchmark signal."],"supporting_citations":[{"why":"Benchmark values for the vector and axial-vector couplings used in the three scenarios.","marker":"[5]"},{"why":"Observed limits from the energetic-jet analysis used as the comparison baseline.","marker":"[25]"},{"why":"Defines the forward proton detector whose acceptance makes the intact-proton tag experimentally available.","marker":"[33]"},{"why":"Provides the complementary forward proton spectrometer used for the same tagging signature.","marker":"[34]"},{"why":"Gives the forward-proton acceptance range 0.015 < xi < 0.15 used for event selection.","marker":"[39]"},{"why":"Earlier study of dark matter in this photon-induced channel and the source of the survival probability S^2 = 0.7.","marker":"[65]"},{"why":"The simplified spin-1 mediator model file used to generate the Z'-mediated interactions.","marker":"[88]"},{"why":"The equivalent photon approximation used to model the quasireal photon flux from the intact proton.","marker":"[89]"}],"fun_headline_variants":["Forward protons tag photon collisions to hunt dark matter","LHC forward-tagging probes dark mediator up to 1.4 TeV","New LHC channel: tag intact protons to see dark matter","Dark matter via photon fusion with forward proton tag","Proton tagging expands LHC dark matter search reach"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The limits stand or fall on the assumption that pileup-produced forward protons can be separated from the signal's own forward proton, even though ordinary jet-plus-missing-energy events with a pileup proton are roughly a thousand times more common and the paper does not specify how the separation would be done.","fun_headline_variants_meta":{"raw":{"variants":["Forward protons tag photon collisions to hunt dark matter","LHC forward-tagging probes dark mediator up to 1.4 TeV","New LHC channel: tag intact protons to see dark matter","Dark matter via photon fusion with forward proton tag","Proton tagging expands LHC dark matter search reach"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000267,"raw_usage":{"total_tokens":1640,"prompt_tokens":999,"completion_tokens":641,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":615,"completion_tokens_details":{"reasoning_tokens":558}},"tokens_in":615,"tokens_out":641,"duration_ms":6452,"temperature":1.0,"reasoning_tokens":558,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:48:19.469408+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A measurement of the rate of jet-plus-missing-energy events that carry a forward proton from pileup, under Run-III conditions with the same $0.05<\\xi<0.15$ cut, would settle the claim: if that rate exceeds the 2305 background events assumed in the 95% confidence calculation, the projected 1.4 TeV mediator exclusion does not survive, and if it can be suppressed below that level, the channel is viable.","supporting_citations":[{"cited_title":"Dark Matter Searches in Jet plus Missing Energy in $\\rm \\gamma p$ collision at CERN LHC","cited_arxiv_id":"1407.5356","evidence_quote":"Earlier study of dark matter in this photon-induced channel and the source of the survival probability S^2 = 0.7."}],"review_version":1}