{"id":"8ea346ae-4ca4-467a-98dc-d2bc5a3ab854","arxiv_id":"2412.19327","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Projected exclusion limits for sbottom masses up to about 1050 GeV at Run-3 and a 100 GeV improvement at HL-LHC, based on scaling an existing ATLAS search.","lead":"This paper simulates signal events for sbottom pair production in a supersymmetric model and projects how far existing ATLAS same-sign lepton searches could exclude new particles at LHC Run-3 and HL-LHC. It is a sensitivity estimate, not a measurement.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Projected reach rests on ad hoc background rescaling and assumed uncertainty reduction; rerunning with Run-2 uncertainties or varied energy factors would test the 1050 GeV and +100 GeV claims.","rationale":"The reader's weakest_assumption points to exactly the same spot I find most load-bearing. The paper is a projection exercise with a clear methodology: it reuses ATLAS signal regions, generates LO signal samples, and applies Eq. (1). The 139 fb^-1 contour is checked against ATLAS, which gives some credibility to the signal modeling. However, that check cannot validate the extrapolation to higher luminosity and energy. The background count at 139 fb^-1 is an ATLAS number, but every future-luminosity number in the paper is obtained by simple scaling of that single number. The energy factors 1.1 and 1.2 are particularly fragile: the same-sign lepton backgrounds contain fake/non-prompt lepton and charge-misidentification contributions that do not scale with the total cross-section or with each other. The reduction of the background uncertainty is also arbitrary; at 3000 fb^-1 the assumed 10% uncertainty dominates the significance. Since the central abstract/conclusion statements give specific mass reaches, those statements should be read as conditional on these assumptions. I do not see an internal inconsistency or a fraudulent step; the limitations are stated openly. A dedicated check with bracketing assumptions would settle how much the numbers move, but the concern is real enough to keep the verdict at CONDITIONAL.","tokens_in":12294,"tokens_out":10594,"duration_ms":98902,"concrete_test":"Recompute the 13.6 TeV/300 fb^-1 and 14 TeV/3000 fb^-1 median exclusion contours with the same DELPHES/SimpleAnalysis pipeline while (a) fixing the background relative uncertainties at the ATLAS Run-2 values (25%/30%) instead of 20%/10%, and (b) varying the energy rescaling factors across 1.0 and 1.3 instead of 1.1/1.2. If either variation moves the 1050 GeV reach or the +100 GeV shift by more than ~50 GeV, the conclusions should label these numbers as scenario-dependent. A stronger check is to rescale the individual background components from Ref. 1 with their own energy and luminosity dependences if public tables allow.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline claims (Section 3/Conclusions: 1050 GeV at 13.6 TeV, 300 fb^-1; generally +100 GeV at 14 TeV) depend on two unvalidated background extrapolations introduced in Section 2. First, the ATLAS Run-2 background counts (6.5 and 7.8 events) are multiplied by 2.16 / 21.60 for luminosity and then by 1.1 (13.6 TeV) and 1.2 (14 TeV) for energy; these energy factors are not derived from the dominant SM background composition (fake/non-prompt leptons, charge flips, VV, ttV), whose energy scaling differs. Second, the relative background uncertainties are reduced from the quoted 25% / 30% to 20% at 300 fb^-1 and 10% at 3000 fb^-1 by assumption. Because Eq. (1) uses both b and sigma, any error in these scalings translates directly into the exclusion contours. The paper is transparent about both simplifications, so this is a limitation rather than a hidden error, but it is load-bearing: the exact numerical reaches are conditional on these choices.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a fast-simulation projection of the LHC search sensitivity for sbottom pair production in an R-parity-conserving simplified model with the decay chain b~1 -> t chi~1^+/- -> t W chi~1^0, using the ATLAS Rpc2L1b and Rpc2L2b same-sign-lepton signal regions defined in JHEP 06 (2020) 046. Signal samples are generated with MadGraph+Pythia at 13, 13.6, and 14 TeV, passed through DELPHES with ATLAS-like object definitions, and analyzed with the SimpleAnalysis framework. Projected exclusion and discovery significances are shown in the (m_b~1, m_chi~1^0) plane for integrated luminosities of 139, 300, and 3000 fb^-1. The headline claims are that at 13.6 TeV with 300 fb^-1 the boosted region could exclude b~1 masses up to 1050 GeV, and that at 14 TeV the Run-2 exclusion limits could generally increase by 100 GeV. The background model is taken from the ATLAS measurement and extrapolated to higher luminosity and energy by multiplicative rescaling, with reduced relative uncertainties by assumption.","tokens_in":12583,"tokens_out":2742,"duration_ms":28179,"significance":"If the projected reaches are robust, the paper provides a useful, quickly reproducible estimate of what the existing ATLAS signal regions could achieve at Run-3 and HL-LHC for this specific simplified model, without new signal-region optimization. The work is transparent about its simplifications, documents acceptance tables and selection step-by-step, and uses standard public tools (MadGraph, Pythia, DELPHES, SimpleAnalysis) and official cross sections. Its main value is as a phenomenological benchmark: it identifies compressed and boosted regions where the current analysis choices are suboptimal and suggests concrete directions for improvement, such as softer lepton thresholds, binned E_T^miss/m_eff fits, and four-lepton channels. The central numerical claims, however, are conditional on an unvalidated background extrapolation and on an assumption about future uncertainty reduction, so the absolute mass reaches should be interpreted with caution rather than as definitive predictions.","major_comments":[{"comment":"The projected limits depend directly on the treatment of the background. The paper multiplies the ATLAS Run-2 background counts by 2.16 (21.60) for luminosity and by 1.1 (1.2) for energy, and reduces the relative background uncertainties from 25%/30% to 20%/10% at 300/3000 fb^-1. These choices enter Eq. (1) through both b and sigma, so any error in them propagates directly into the exclusion contours. The paper is explicit that these are assumptions, but it does not justify the energy factors from the dominant background components (fake/non-prompt leptons, charge flips, VV, ttV), whose production and fake-rate scalings are different. I request a robustness study: repeat the limit calculation with the Run-2 relative uncertainties unchanged, with energy factors varied over a plausible range, and with an alternative luminosity scaling that does not assume the background is fully statistical. If the headline 1050 GeV and +100 GeV changes are stable under these variations, that should be shown; if they are not, the conclusions should be reworded accordingly.","section":"Section 2, background extrapolation paragraph"},{"comment":"The 14 TeV projections are obtained from 13.6 TeV event samples normalized to 14 TeV production cross sections, with the justification that 'this is a reasonable approach, as confirmed with the 13 TeV and 13.6 TeV samples.' No quantitative comparison of acceptance or kinematic distributions between 13 and 13.6 TeV samples is shown anywhere in the paper. Since the visible energy available in the sbottom decay chain shifts with centre-of-mass energy, acceptance differences in the high-m_eff, high-E_T^miss signal regions are not guaranteed to be negligible. Please provide a validation plot or table comparing acceptance and signal yields at 13 and 13.6 TeV for representative mass points, and estimate the systematic uncertainty introduced by using 13.6 TeV samples for 14 TeV projections.","section":"Section 3, Fig. 6 and surrounding text"},{"comment":"The statement that the limits are obtained 'without optimization' is slightly misleading, because the best signal region (Rpc2L1b or Rpc2L2b) is chosen per mass point based on the largest significance Z. This is a mild form of optimization, and while it is conventional in such projection studies, it should be acknowledged explicitly as such, since it can inflate the apparent reach compared to a fixed signal-region strategy. The effect is likely small because only two regions are considered, but the text should clarify this.","section":"Section 3, best-signal-region choice"}],"minor_comments":[{"comment":"The manuscript contains several typographical and grammatical errors, including 'the exclusion limits in the boosted could reach' (conclusions) and 'the discovery potential' used without a clear object. The text should be carefully proofread.","section":"General"},{"comment":"The sign in Eq. (1) is written with a leading '+/-' in front of the square root, which is unconventional; the standard asymptotic significance formula has no leading sign ambiguity. Clarify that Z is taken as positive for an excess and negative for a deficit, or cite the exact equation from Ref. [21].","section":"Section 1, Eq. (1)"},{"comment":"The two pre-selection lists for Rpc2L1b and Rpc2L2b are presented with bullets that are visually identical; please add labels or separate them more clearly, as in the current format the reader must infer which list belongs to which signal region.","section":"Section 2, signal region definitions"},{"comment":"The captions of Figs. 4-6 state that the z axis shows signal significance but do not explain how the exclusion contour is extracted from the significance map, beyond mentioning Z=1.64 in the text. A sentence on the contour-finding procedure would improve reproducibility.","section":"Figure captions"},{"comment":"Reference [15] cites an online manual for MLM matching rather than the original MLM publication; consider adding the canonical reference for completeness.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper is a single-author phenomenological projection that is transparent about its limitations. The central issue is not that the assumptions are hidden, but that the two most important numerical claims rest on unvalidated background extrapolations and on an assumed reduction of systematic uncertainties. With a modest amount of additional work (robustness scans and a 13 vs 13.6 TeV validation) the paper could become a reliable benchmark. I see no reason to doubt the basic methodology, but the current version does not yet establish the robustness needed for the claimed mass reaches to be quoted as definitive."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis paper is a projection, not a new measurement. The author maps the ATLAS Run-2 sbottom search in same-sign leptons (JHEP06(2020)046) onto LHC Run-3 and HL-LHC conditions, using fresh MadGraph+Pythia+Delphes signal samples at 13, 13.6, and 14 TeV. The output is a set of projected exclusion and discovery significances in the sbottom–neutralino mass plane, with headline numbers of about 1050 GeV reach at 13.6 TeV with 300 fb^-1 and a general +100 GeV gain at 14 TeV. What is genuinely useful is that the signal regions, object definitions, and background inputs come straight from the ATLAS paper, so the projection is easy to compare against the original search.\n\nThe author is transparent about the simplifications, and that transparency is the paper's main virtue. The background counts from ATLAS are scaled linearly with luminosity, multiplied by hand-assigned energy factors (1.1 for 13.6 TeV, 1.2 for 14 TeV), and the relative background uncertainties are reduced from 25%/30% to 20% at 300 fb^-1 and 10% at 3000 fb^-1. These assumptions sit directly in the significance formula, so they determine the exact contour locations. The stress-test note is right: the numbers are conditional on these choices. A plot validating the 13.6-to-14 TeV normalization claim would also help; the paper says it's confirmed but does not show it.\n\nNone of this breaks the paper. The signal simulation is standard, the significances are computed with a standard formula, and the per-mass-point choice of the better signal region is driven by signal acceptance, not by background fluctuations. The limitations are stated, not hidden. But they are load-bearing.\n\nFor whom: LHC search planners and phenomenologists who want a quick, honest estimate of what the existing Run-2 signal regions could do with more data. It is not a new physics result. It does deserve a serious referee, mostly to push on the background-scaling sensitivity and request a validation plot. I would not cite it in my own work, but I would put it in front of a reading group for a reality check on upcoming SUSY search sensitivities.\n\nBest.","headline":"Transparent projection of an existing ATLAS sbottom search to Run-3 and HL-LHC; the reach numbers are conditional on untested background scaling assumptions, but the paper is honest about them.","tokens_in":13045,"tokens_out":3639,"would_cite":false,"duration_ms":30651,"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":"Projection: LHC Run-3 sbottom searches can exclude masses up to 1050 GeV.","keywords":["sbottom pair production","same-sign leptons","supersymmetry","LHC Run-3","HL-LHC","signal region projections","DELPHES fast simulation","chargino-mediated decay"],"falsifier":"Run the same Rpc2L1b and Rpc2L2b selections on the first 300 fb$^{-1}$ of real 13.6 TeV ATLAS data: if the observed background exceeds the scaled prediction, or if the resulting 95% CL exclusion contour fails to reach $\\tilde b_1$ masses near 1050 GeV in the boosted region, the projection is falsified. A second check is to measure whether the background uncertainty actually drops from 25--30% to about 20% at 300 fb$^{-1}$.","tokens_in":12127,"feed_emoji":"⚛️","tokens_out":4784,"duration_ms":38243,"temperature":0.7,"pith_summary":"This paper projects how far the LHC's next data-taking periods could push the search for sbottom squarks in a specific supersymmetric decay chain: $\\tilde b_1 \\to t \\tilde\\chi_1^\\pm$, followed by $\\tilde\\chi_1^\\pm \\to W \\tilde\\chi_1^0$, each at 100% branching ratio. The author reuses the ATLAS Run-2 signal regions designed for same-sign dileptons plus jets, simulates signal samples with fast detector simulation, and scales the ATLAS background estimates to higher luminosity and energy. The headline projection is that at 13.6 TeV with 300 fb$^{-1}$, sbottom masses up to 1050 GeV could be excluded in the boosted region, and that at 14 TeV the Run-2 exclusion reach could grow by about 100 GeV without any signal-region optimization. A sympathetic reader would care because these are concrete, testable benchmarks for whether the upcoming LHC runs can discover or rule out this motivated SUSY scenario.","feed_headline":"Run-3 sbottom search can exclude up to 1050 GeV","feed_subtitle":"Projections from ATLAS same-sign lepton regions show HL-LHC extending Run-2 limits by 100 GeV.","key_machinery":"The machinery is the pair of ATLAS signal regions Rpc2L1b and Rpc2L2b (final states with at least two same-sign leptons, at least one or two $b$-tagged jets, at least six jets, and cuts on $E_T^{\\text{miss}}/m_\\text{eff}$), implemented in the SimpleAnalysis framework on DELPHES fast-simulated events. The reach estimate then comes from the signal significance formula of Eq. (1), with background counts taken from the ATLAS publication and scaled linearly with luminosity and by factors of 1.1 (13.6 TeV) and 1.2 (14 TeV) for energy, while background uncertainties are assumed to shrink from 25--30% to 20--10% at higher luminosity.","core_discovery":"The paper's central claim is that the existing ATLAS Rpc2L1b and Rpc2L2b signal regions retain enough sensitivity that the upcoming runs can substantially extend sbottom exclusion without redesign. Concretely, at $\\sqrt{s}=13.6$ TeV with 300 fb$^{-1}$, the boosted region could exclude $\\tilde b_1$ masses up to 1050 GeV, the compressed region up to 850 GeV, and the intermediate region from 950 to 980 GeV. At $\\sqrt{s}=14$ TeV, the Run-2 exclusion limits could generally increase by 100 GeV, and signal significances around 5 or above in the 950--1050 GeV sbottom mass window would make discovery possible if the model is realized in nature.","pith_inferences":["The linear background scaling and simplified energy factors are the fragile part of the extrapolation; a direct data-driven background estimate at 13.6 TeV early in Run-3 would provide a much firmer projection.","The same projection recipe could be applied to stop or gluino simplified models with chargino-mediated decays, giving quick first estimates of their Run-3 and HL-LHC reach.","If pileup degrades $b$-tagging or lepton isolation more than assumed, the balance between the one-$b$-tag and two-$b$-tag regions would shift, changing which region drives the limit.","The paper's sensitivity maps suggest that a binned fit in $E_T^{\\text{miss}}/m_\\text{eff}$, rather than a single cut, is the most promising cheap upgrade for the boosted region."],"forward_implications":["At 13.6 TeV with 300 fb$^{-1}$, without changing the ATLAS signal regions, the boosted-region sbottom exclusion could reach 1050 GeV.","At the HL-LHC, the Run-2 sbottom exclusion limits could rise by about 100 GeV, with additional sensitivity coming from reduced background uncertainties.","If the sbottom-to-chargino decay chain is real, HL-LHC data could yield discovery-level significance ($Z \\ge 5$) for sbottom masses between 950 and 1050 GeV.","Dedicated optimization of the signal regions (softer leptons, 4-lepton channels, binned or machine-learning selections) could extend reach beyond these unmodified-region projections."],"supporting_citations":[{"why":"Provides the signal region definitions, object selections, and the background estimates (6.5 and 7.8 events) that the projections scale.","marker":"Ref. 1"},{"why":"Supplies the production cross sections used to normalize signal samples at 13, 13.6, and 14 TeV.","marker":"Ref. 8"},{"why":"The fast detector simulation framework used to model the ATLAS response to signal events.","marker":"Ref. 16"},{"why":"Provides the asymptotic significance formula (Eq. 1) used to compute Z and set exclusion and discovery thresholds.","marker":"Ref. 21"},{"why":"The truth-level analysis framework used to implement the ATLAS event selection.","marker":"Ref. 20"},{"why":"An alternative same-sign lepton search with different LSP mass choices, mentioned as comparison and context for the model.","marker":"Ref. 11"}],"fun_headline_variants":["Sbottom search can exclude up to 1.05 TeV at Run-3","HL-LHC extends sbottom exclusion by 100 GeV","Same-sign leptons unveil sbottom search potential","Projected sbottom limits reach 1050 GeV with 300 fb^-1","Discovery window for sbottom emerges at 950–1050 GeV"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The projected limits rely on the assumption that the ATLAS background counts of 6.5 and 7.8 events scale linearly with luminosity and with the simple 1.1 and 1.2 energy factors, while their relative uncertainties fall as assumed; if background composition or systematics change with pileup and energy, the limits shift.","fun_headline_variants_meta":{"raw":{"variants":["Sbottom search can exclude up to 1.05 TeV at Run-3","HL-LHC extends sbottom exclusion by 100 GeV","Same-sign leptons unveil sbottom search potential","Projected sbottom limits reach 1050 GeV with 300 fb^-1","Discovery window for sbottom emerges at 950–1050 GeV"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000722,"raw_usage":{"total_tokens":3251,"prompt_tokens":967,"completion_tokens":2284,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":583,"completion_tokens_details":{"reasoning_tokens":2191}},"tokens_in":583,"tokens_out":2284,"duration_ms":15828,"temperature":1.0,"reasoning_tokens":2191,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T00:41:21.532622+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same Rpc2L1b and Rpc2L2b selections on the first 300 fb$^{-1}$ of real 13.6 TeV ATLAS data: if the observed background exceeds the scaled prediction, or if the resulting 95% CL exclusion contour fails to reach $\\tilde b_1$ masses near 1050 GeV in the boosted region, the projection is falsified. A second check is to measure whether the background uncertainty actually drops from 25--30% to about 20% at 300 fb$^{-1}$.","supporting_citations":[],"review_version":1}