{"id":"a8ae4080-af24-4294-a6b7-6a2507c58b56","arxiv_id":"1908.11350","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Projected HE-LHC searches could rule out neutralino dark matter up to 2.6, 1.7, and 0.8 TeV in gluino, stop, and wino coannihilation scenarios, but not in stau coannihilation.","lead":"This paper simulates how well a proposed 27 TeV upgrade of the LHC, the HE-LHC, could detect dark matter particles in supersymmetric models where dark matter co-annihilates with a heavier partner particle. It finds that gluino, stop, and wino co-annihilation scenarios could be probed up to neutralino masses of 2.6, 1.7, and 0.8 TeV, while stau co-annihilation would remain out of reach.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quoted 2-sigma reach assumes signal-only systematics; the paper admits this, yet the mass limits are reported without any systematic uncertainty budget, so the 2.6/1.7/0.8 TeV numbers are optimistic.","rationale":"The reader's weakest_assumption correctly identifies the systematics omission as the main issue. My independent read of the paper confirms that Z = S/sqrt(B) is used throughout (Section 3 introduction) and that the authors explicitly disclaim systematics in Section 3.4. The paper is a well-structured phenomenological projection: it uses standard tools (MG5_aMC, Pythia8, Delphes, CheckMATE), imposes relic density and LHC constraints, and gives a clear qualitative ranking (gluino > stop > wino > stau). The concrete numerical limits, however, are not robust to background systematics. I do not think this warrants a change in verdict from CONDITIONAL, because the reader already flags the issue and the paper itself admits the limitation; the conditional acceptance is appropriate. I see no internal inconsistency, no computational error, and no unsupported claim beyond the known systematics caveat, so I agree with the reader's assessment. The proposed test (adding a 10% systematic uncertainty and recomputing) would turn the reader's qualitative concern into a quantitative estimate of how much the quoted reach would shrink.","tokens_in":18763,"tokens_out":1693,"duration_ms":14627,"concrete_test":"Recompute the 2-sigma exclusion curves for the gluino, stop, and wino scenarios at L = 15 ab^-1 after adding a flat 10% background systematic uncertainty to each signal region (sigma_bkg = sqrt(B + (0.1*B)^2)). If the new 2-sigma mass limits drop by more than ~15% from 2.6/1.7/0.8 TeV, the headline numbers should be restated as systematics-limited upper bounds. Additionally, validate the Delphes/CheckMATE fast simulation by reproducing the ATLAS 13 TeV 36.1 fb^-1 limits in Fig. 1; if the recast does not match the published limits within 20%, the HE-LHC projections inherit that calibration error.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline mass limits are derived from Z = S/sqrt(B) with no systematic uncertainties, as stated in Section 3 (\"We evaluate the statistical significance with the formula Z = S/sqrt(B)\") and acknowledged in the final paragraph of Section 3.4 (\"the statistical significance will get degraded when systematic uncertainties are taken into account\"). For a 27 TeV HE-LHC with 15 ab^-1, background yields in the multijet and monojet searches are enormous, and even a 10-20% background systematic uncertainty would shift the quoted 2-sigma exclusions by several hundred GeV, especially for the gluino and stop scenarios where the signal is suppressed by small mass splittings and relies on high-ETmiss tails. The paper provides no systematic uncertainty budget, no validation of the Delphes/CheckMATE fast simulation against the ATLAS/CMS public results at 13 TeV, and no discussion of how pile-up at 27 TeV (200 interactions per bunch crossing) would degrade jet reconstruction, tau-tagging, and ETmiss resolution. Since the central claim is a quantitative reach statement, the absence of a systematics treatment makes the quoted mass limits optimistic upper bounds rather than robust projections.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the reach of a 27 TeV HE-LHC with 15 ab^-1 of integrated luminosity for neutralino dark matter in bino-gluino, bino-stop, bino-wino, and bino-stau coannihilation scenarios within a simplified MSSM. The authors first select model points satisfying the Planck relic density, the measured Higgs mass, vacuum stability, and current LHC constraints using MicrOMEGAs and SUSY-HIT. They then simulate signal and background events with MG5_aMC@NLO, Pythia8, Delphes3.4.1, and CheckMATE, defining signal regions for multijet + missing transverse energy, monojet, soft same-flavor dilepton + missing transverse energy, and monojet + hadronic tau. Using a simple significance Z = S/sqrt(B), they report 2 sigma exclusion reaches of 2.6 TeV, 1.7 TeV, and 0.8 TeV for the neutralino mass in gluino, stop, and wino coannihilations, respectively, and no sensitivity in stau coannihilation. The paper acknowledges in the final paragraph of Section 3.4 that systematic uncertainties and pile-up are not modeled and would degrade the quoted significances.","tokens_in":19072,"tokens_out":6350,"duration_ms":66345,"significance":"If the quoted reach numbers were robust, the paper would provide a useful quantitative guide for HE-LHC SUSY searches in coannihilation scenarios, extending existing 13 TeV limits and identifying stau coannihilation as the most difficult channel. The methodology is a standard phenomenological projection chain using publicly available Monte Carlo tools, with a clear scan and constraint procedure for each scenario. The qualitative hierarchy of reach (gluino > stop > wino >> stau) is likely to survive more careful treatment, and the explicit caveats about systematics and pile-up are honest. However, the central quantitative claim is presented as exclusion reaches, and the analysis is statistical-only, with no systematic uncertainty budget, no high-pile-up modeling, and LO-only cross sections. The paper therefore needs revision before the specific TeV numbers can be taken as reliable projections.","major_comments":[{"comment":"The headline reach numbers, such as the 2.6, 1.7, and 0.8 TeV 2-sigma exclusions quoted in the abstract and conclusions, are computed with Z = S/sqrt(B) and no systematic uncertainties. The paper itself states in the final paragraph of Section 3.4 that 'the statistical significance will get degraded when systematic uncertainties are taken into account.' For the multijet and monojet searches at 27 TeV with 15 ab^-1, the background yields are large, and a moderate background systematic uncertainty of 10-20% can shift a 2-sigma exclusion by several hundred GeV in mass. Since the central claim of the paper is a set of quantitative mass limits, the analysis should either include a nuisance-parameter treatment with representative systematic uncertainties, or the abstract and conclusions should explicitly label all quoted reaches as statistical-only sensitivities rather than as exclusion projections.","section":"Abstract/Conclusions; Section 3, Z = S/sqrt(B)"},{"comment":"All signal and background cross sections are evaluated at leading order with MG5_aMC@NLO. For gluino pair production, which is gg-initiated, NLO QCD corrections and scale uncertainties are known to be sizable; stop pair production also receives non-negligible QCD corrections. Since the quoted reach numbers depend directly on signal and background normalizations, the absence of k-factors or scale/PDF uncertainty bands leaves the mass limits with an unquantified normalization error. At minimum, the authors should estimate the impact of NLO corrections to the signal cross sections, or state as an additional caveat that all reach numbers are LO-normalized.","section":"Section 3, first paragraph"},{"comment":"The stau analysis assumes a flat 60% efficiency for hadronic tau tagging for taus with pT between 15 and 35 GeV. Realistic tau tagging efficiencies at such low pT are substantially lower and depend strongly on pT, and they are further degraded by the high pile-up expected at HE-LHC. The stau channel conclusion of 'no sensitivity' is one of the paper's quantitative results, so this assumption should be varied over a plausible range or replaced by a pT-dependent efficiency curve to demonstrate that the conclusion is robust.","section":"Section 3.4, tau tagging bullet"},{"comment":"The paper correctly notes that pile-up effects are beyond its scope, but pile-up at 27 TeV (O(200) interactions per bunch crossing) directly affects the soft-lepton, soft-tau, and E_T^miss observables that drive all four analyses. The manuscript does not need a full detector simulation, but a quantitative statement of how the quoted reaches change under plausible E_T^miss resolution degradation or reconstruction efficiency losses is necessary if the abstract's 'excluded' language is retained. Without this, the reported mass limits are upper bounds on statistical-only sensitivity rather than robust exclusion projections.","section":"Section 3.4, final paragraph"}],"minor_comments":[{"comment":"The scan ranges are formatted badly, e.g., '100 GeV < M 1, 3< 3 TeV' should presumably read '100 GeV < M1, M3 < 3 TeV', with similar problems in the stop, wino, and stau scan ranges. Please fix the notation.","section":"Section 2, scan ranges"},{"comment":"There are several typographical errors: 'enenrgy' in the multijet event-selection bullet, 'processs' in the gluino paragraph, and 'Events franction' in the Figure 8 caption. These should be corrected.","section":"Section 3.1 and Figure 8 captions"},{"comment":"Reference [92] is incomplete; it lists only the title and collaboration and lacks the journal, volume, article number, and arXiv identifier.","section":"Reference [92]"},{"comment":"The PDF set is written as 'NN23LO1'; this should be 'NNPDF23LO1' or another complete PDF set name to be unambiguous.","section":"Section 3, Monte Carlo setup"},{"comment":"The text refers to a transverse mass mT(l, nu_l) in the motivation, but the selection criteria use mT(l1, E_T^miss). Please define the variable used in the selection explicitly or reconcile the notation.","section":"Section 3.3, transverse mass definition"}],"recommendation":"major_revision","confidential_remarks":"The paper is a standard phenomenological projection and is within the scope of JHEP. The qualitative conclusions are likely robust, but the quantitative reach numbers are presented without a systematics or pile-up treatment and with LO-only cross sections, so the revision should focus on either adding those estimates or softening the claims consistently throughout the abstract, conclusions, and main text."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper does a consistent scan of four neutralino coannihilation scenarios and gives 27 TeV LHC reach projections with 15 ab^-1. The headline numbers—2.6 TeV gluino, 1.7 TeV stop, 0.8 TeV wino at 2σ, and no stau sensitivity—come from real Monte Carlo work, not a back-of-envelope. I think the qualitative story is solid.\n\nWhat is new: earlier papers looked at individual coannihilation scenarios at 13/14 TeV, and some HE-LHC projections exist, but I do not know of a single paper that treats gluino, stop, wino, and stau coannihilation in the same simplified MSSM framework, imposes relic density and LHC constraints consistently, and then projects to 27 TeV with the same detector simulation. That makes the comparison between scenarios meaningful. The authors use MicrOMEGAs for relic density and CheckMATE/Delphes for collider analysis, which is standard, and they define signal regions carefully enough that the analysis is basically reproducible from the text. They also correctly note that Sommerfeld and bound-state effects would not change the collider reach, which shows they know where the relic-density calculation is limited and why it does not bite here.\n\nThe soft spots, in order. First, the significance is Z = S/sqrt(B) with no systematic uncertainty. The authors admit this in the final paragraph of Section 3.4, but the abstract and conclusions still present the 2σ numbers as flat facts. For a 27 TeV machine with 15 ab^-1, backgrounds are huge, and even a 10–20% background systematic would pull the gluino and stop limits down by several hundred GeV. Treat the quoted numbers as optimistic upper bounds. Second, cross sections are LO only, and there is no validation of the fast simulation against 13 TeV public results beyond checking signal regions. That is common in this literature, but it limits precision. Third, τ-tagging is a flat 60% and pile-up is not modeled; the authors explicitly leave pile-up for future work, which is honest but makes the stau non-sensitivity claim the least certain of the four. That said, the stau conclusion is unlikely to flip—the cross section is small—but the exact wording depends on a simple tau model.\n\nNone of this changes the central ordering: gluino > stop > wino >> stau, with rough mass reaches as stated. The systematics critique should not be read as “the paper is worthless”; it is a solid phenomenological projection, honestly written. Who is it for: people planning SUSY searches at a 27 TeV collider, and anyone working on compressed spectra or coannihilation. It deserves a serious referee; the main referee ask should be a systematics sensitivity check, or at least a statement of how much the limits move with a few percent background uncertainty. I would cite it if I wrote about HE-LHC SUSY reach.","headline":"Competent HE-LHC projection for four coannihilation scenarios; the mass limits are optimistic because they are statistical-only, but the qualitative ordering is probably right.","tokens_in":19529,"tokens_out":2367,"would_cite":true,"duration_ms":25481,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["95.35.+d","12.60.Jv"],"model":"deepseek-v4-flash","headline":"The paper projects that a 27 TeV HE-LHC with 15 ab^-1 can exclude neutralino dark matter up to 2.6 TeV in gluino coannihilation, 1.7 TeV in stop coannihilation, and 0.8 TeV in wino coannihilation at 2 sigma, while stau coannihilation…","keywords":["neutralino dark matter","coannihilation","HE-LHC","compressed supersymmetry","gluino coannihilation","stop coannihilation","wino coannihilation","stau coannihilation"],"falsifier":"Run one benchmark point in each channel through a full detector simulation with realistic tagging efficiencies and add a 10 percent systematic uncertainty on background yields; if the significance drops below 2 $\\sigma$ for a gluino around $m_{\\tilde g}=1.6$ TeV and a neutralino around $m_{\\tilde\\chi_1^0}=1.5$ TeV, the claimed 2.6 TeV reach is not robust.","tokens_in":18570,"feed_emoji":"⚛️","tokens_out":8245,"duration_ms":72225,"temperature":0.7,"pith_summary":"The paper tries to establish that a 27 TeV High-Energy LHC with 15 $ab^{-1}$ of data could test most of the cosmologically allowed coannihilation windows for neutralino dark matter. In supersymmetric models, the relic abundance of a bino-like neutralino can be brought into agreement with observation when it is nearly degenerate with a gluino, stop, wino, or stau; the near degeneracy makes the decay products soft and the searches difficult. By simulating four dedicated search channels, the authors find 2-$\\sigma$ exclusion reaches of 2.6 TeV, 1.7 TeV, and 0.8 TeV for gluino, stop, and wino coannihilation respectively, while stau coannihilation gives no reach. This matters because coannihilation is a leading way to make neutralino dark matter viable, and a positive result would let a collider discover the partner particle and thereby measure the dark matter mass.","feed_headline":"HE-LHC could exclude neutralino dark matter to 2.6 TeV","feed_subtitle":"Gluino, stop and wino coannihilation windows are in reach at 27 TeV; stau coannihilation remains invisible.","key_machinery":"The machinery is the coannihilation strip together with ISR-boosted compressed searches. The strip is the set of points in the neutralino-mass versus partner-mass plane where the relic density is satisfied because a nearly degenerate partner, with mass splitting from a few GeV to about 100 GeV, participates in the freeze-out. To see the soft final states, each search requires a hard jet from initial-state radiation and then defines signal regions in variables such as the effective mass, missing transverse energy, dilepton invariant mass, and transverse mass; significance is evaluated as $Z=S/\\sqrt{B}$ with background yields from a leading-order simulation.","core_discovery":"The paper's central claim is a set of projected 2-$\\sigma$ exclusion limits for the neutralino dark matter mass in four coannihilation scenarios at the HE-LHC: 2.6 TeV in gluino coannihilation via multijets plus missing transverse energy, 1.7 TeV in stop coannihilation via monojets, and 0.8 TeV in wino coannihilation via a soft same-flavour lepton pair plus missing transverse energy, all at 15 $ab^{-1}$; stau coannihilation via a monojet plus one hadronic tau remains below 2 $\\sigma$ even at that luminosity. The limits follow from first restricting each scenario to samples that reproduce the observed dark matter relic density within 2 $\\sigma$, then simulating the compressed final states with initial-state-radiation jets to boost the soft system. The authors state that the stau failure is due to the small direct stau pair-production cross section and the low tagging efficiency for soft taus from stau decay.","pith_inferences":["Inference: Because the quoted significance neglects systematic uncertainties, the real HE-LHC reach could be materially lower; a robust projection should fold in correlated background systematics and pile-up effects, which the paper leaves to future work.","Inference: The stau blind spot means that if future data select stau coannihilation as the only surviving neutralino dark matter window, HE-LHC alone cannot close it; complementary probes such as lepton colliders, long-lived-particle searches, or improved soft-tau tagging would be needed.","Inference: The same ISR-boosted search logic could be applied to the HL-LHC at 14 TeV to give a lower but still meaningful reach, and to a 100 TeV collider to push the same strips to higher masses; neither extrapolation is made in the paper."],"forward_implications":["At 27 TeV with 15 ab^-1, the gluino coannihilation strip can be excluded up to a neutralino mass of about 2.6 TeV at 2 sigma, with a 5-sigma discovery reach of about 2.2 TeV.","In stop coannihilation, the monojet search can exclude neutralino masses up to about 1.7 TeV at 2 sigma, with a 5-sigma reach below about 1.4 TeV at 15 ab^-1.","The soft-dilepton search for wino coannihilation extends the neutralino exclusion from roughly 180 GeV at the current LHC to about 560 GeV at 300 fb^-1 and 0.8 TeV at 15 ab^-1, both at 2 sigma.","Stau coannihilation remains out of reach at the HE-LHC: even at 15 ab^-1, no sample reaches 2 sigma, because stau pair production is weak and the soft tau from stau decay is poorly tagged.","If no excess is seen, the combination of these channels would rule out most of the relic-density-allowed gluino, stop, and wino coannihilation parameter space at the HE-LHC."],"supporting_citations":[{"why":"Supplies the effective coannihilation cross-section formula that defines how a nearly degenerate partner changes the relic abundance.","marker":"[50]"},{"why":"Computes the dark matter relic density for each scanned sample, providing the central constraint that selects allowed coannihilation points.","marker":"[51]"},{"why":"Provides the 2-sigma Planck relic-density range used to define cosmologically allowed samples.","marker":"[53]"},{"why":"The LHC gluino search whose null results set the current lower bound on gluino mass and whose signal regions are re-optimized for the HE-LHC.","marker":"[58]"},{"why":"The LHC monojet search used as the baseline for the stop coannihilation channel and for setting current stop limits.","marker":"[59]"},{"why":"The LHC soft-dilepton compressed-search analysis that defines the current wino coannihilation limit and is adapted to the HE-LHC.","marker":"[22]"},{"why":"The LHC stau search that sets the current stau mass limit and motivates the single-tau plus ISR-jet channel studied here.","marker":"[92]"},{"why":"Generates the leading-order signal and background events used for all projected significance estimates.","marker":"[80]"}],"fun_headline_variants":["Neutralino DM exclusion tops 2.6 TeV at HE-LHC","HE-LHC: stau coannihilation remains beyond reach","Wino neutralino DM down to 0.8 TeV at HE-LHC","HE-LHC excludes neutralino DM up to 2.6 TeV","Gluino coannihilation: neutralino DM limit 2.6 TeV"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the background can be predicted well enough for $Z=S/\\sqrt{B}$ to describe the real search; if systematic uncertainties in the background or detector effects are substantial, the quoted 2-$\\sigma$ reaches shrink.","fun_headline_variants_meta":{"raw":{"variants":["Neutralino DM exclusion tops 2.6 TeV at HE-LHC","HE-LHC: stau coannihilation remains beyond reach","Wino neutralino DM down to 0.8 TeV at HE-LHC","HE-LHC excludes neutralino DM up to 2.6 TeV","Gluino coannihilation: neutralino DM limit 2.6 TeV"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000662,"raw_usage":{"total_tokens":3063,"prompt_tokens":1021,"completion_tokens":2042,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":637,"completion_tokens_details":{"reasoning_tokens":1939}},"tokens_in":637,"tokens_out":2042,"duration_ms":14663,"temperature":1.0,"reasoning_tokens":1939,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:17:11.228912+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run one benchmark point in each channel through a full detector simulation with realistic tagging efficiencies and add a 10 percent systematic uncertainty on background yields; if the significance drops below 2 $\\sigma$ for a gluino around $m_{\\tilde g}=1.6$ TeV and a neutralino around $m_{\\tilde\\chi_1^0}=1.5$ TeV, the claimed 2.6 TeV reach is not robust.","supporting_citations":[{"cited_title":"The Coannihilation Codex","cited_arxiv_id":"1510.03434","evidence_quote":"Supplies the effective coannihilation cross-section formula that defines how a nearly degenerate partner changes the relic abundance."},{"cited_title":"Search for electroweak production of supersymmetric states in scenarios with compressed mass spectra at $\\sqrt{s}=13$ TeV with the ATLAS detector","cited_arxiv_id":"1712.08119","evidence_quote":"The LHC soft-dilepton compressed-search analysis that defines the current wino coannihilation limit and is adapted to the HE-LHC."}],"review_version":1}