{"id":"0be61730-7171-417e-aabe-78ac5217101e","arxiv_id":"1012.3375","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"No microscopic black holes appeared in 35 inverse picobarns of 7 TeV CMS data, giving 95% confidence lower limits of 3.5 to 4.5 TeV on the minimum black hole mass in large-extra-dimension models.","lead":"CMS searched 35 inverse picobarns of 7 TeV proton collisions for microscopic black holes, which models with extra spatial dimensions predict should appear as high-multiplicity, high-transverse-energy bursts. No excess appeared, and the resulting mass limits are the first direct constraints on black hole production at a collider.","discovery_kind":"extension","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"The N≥4,5 background is extrapolated from N=2,3 fits; the authors price this in with a 6–125% systematic, but the shape is validated only at adjacent multiplicities, so the N≥5 extrapolation is the least-tested link in the mass limit.","rationale":"The reader's weakest_assumption is the right one: the S_T shape invariance across multiplicity is the load-bearing assumption for the N≥4,5 background, and it is the least directly tested part of the analysis. I agree with the diagnosis and with the ACCEPT verdict. The reason the verdict does not move is that the authors do not hide the assumption: they state it in the paragraph beginning 'For QCD events, S_T is almost completely determined by the hard 2→2 parton scattering process,' they validate it with MC and with the N=2 vs N=3 agreement, and they carry a 6–125% background systematic into the limit calculation. The quoted σ95 values already marginalize over that uncertainty, so the concern is priced in rather than unaddressed. Two caveats sharpen but do not overturn the result. First, the classifier's 'Limitations the authors state: (none found)' is inaccurate: the paper explicitly flags the model dependence of the cross section ('The exact cross section cannot be calculated without knowledge of the underlying theory of quantum gravity'), the no-energy-loss assumption ('We assume that no parton-collision energy is lost in gravitational shock waves'), and the correlated background uncertainty. These are acknowledged, so they do not count as new findings. Second, the M_D convention is a real model dependence: the parenthetical that the Dimopoulos-Landsberg definition suppresses the cross section by up to 9.29 for n=6 implies the 4.5 TeV n=6 limit would move down by several hundred GeV under that convention. The paper is transparent about this, and the abstract's 'variety of parameters' is defensible, but a reader should not take 4.5 TeV as convention-independent. Neither caveat changes the verdict: the analysis is internally consistent, the background is data-driven, the signal is external, and the model-independent limits in Fig. 5 outlive the ADD benchmarks. The concrete test above is the one check that would tell us whether the 6–125% band is adequately sized at N≥5.","tokens_in":29314,"tokens_out":12122,"duration_ms":216936,"concrete_test":"Compare the N≥5 data to the N=2,3-derived shape in the signal-free window S_T=700–1000 GeV (above the N×50 GeV turn-on, below the 1000–1100 GeV normalization window), where Fig. 2c currently shows no points. Compute the data/prediction ratio and its uncertainty. If the ratio is consistent with unity within the assigned systematic, the extrapolation is validated; if it deviates by more than the systematic, refit the N≥5 shape directly in 700–1000 GeV and recompute σ95 at S_T^min=2.4–2.9 TeV. A shift in σ95 larger than the quoted 0.11–0.15 pb would mean the shape extrapolation, not statistics, sets the n=6 mass limit.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (no excess; 3.5–4.5 TeV limits) rests on a data-derived QCD background for N≥4 and N≥5 that is never fit directly. The ansatz P0(1+x)^P1/(x^P2+P3 log x) is fitted to N=2 and N=3 data in S_T=600–1100 GeV and then normalized to S_T=1000–1100 GeV; the same shape is applied to N≥4,5 (Figs. 2b,c). The systematic band (6–125%) is derived from alternate ansatz functions and the N=2 vs N=3 difference. That is a two-point check at adjacent multiplicities; it does not probe the N=3→N=5 gap, and the normalization region for N≥5 may be statistics-limited, so the 4–12% normalization uncertainty could be underestimated there. This matters because the mass limits are set at S_T^min=2.4–2.9 TeV (N≥5), where the expected background is 0.5–2 events with ~70–100% uncertainty; if the true N≥5 shape is harder than the N=2,3 fit, the expected background rises and σ95 weakens. The data shown in Fig. 2c above 1100 GeV does follow the curve, which is partial evidence the shape is right, but that is also the signal region, so it cannot cleanly separate a shape error from a signal deficit.","agreement_with_reader":"agree"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The CMS Collaboration reports a search for microscopic black hole production and decay in 7 TeV pp collisions using 34.7 ± 3.8 pb^-1. Events are selected on total transverse energy S_T, defined as the scalar sum of E_T of jets, electrons, photons and muons above 50 GeV, plus missing transverse energy above 50 GeV, and are binned in the number of objects N. The dominant QCD multijet background is extracted from data by fitting the S_T distribution of the N = 2 and N = 3 samples in the range 600–1100 GeV with a parametric ansatz, checking the fit against two alternate functional forms, and extrapolating the resulting shape to the N ≥ 3, 4, 5 selections. No excess is observed. The paper sets 95% CL upper limits on the black hole production cross section as a function of M_BH^min for three ADD benchmark points, converts these into lower limits on M_BH^min of 3.5, 4.2 and 4.5 TeV (expected 3.2, 4.0, 4.5 TeV), and presents model-independent limits on cross section times acceptance for high-S_T, high-multiplicity final states. The stated novelty is that these are the first direct limits on black hole production at a particle accelerator.","tokens_in":29540,"tokens_out":7063,"duration_ms":124329,"significance":"If the result holds, it is a genuine and durable contribution: the first collider constraint on a striking low-scale-gravity signature, delivered on 35 pb^-1 of early LHC data. Several features deserve explicit credit. The background is derived from data rather than from simulation, which is the right choice for a QCD-dominated search and removes the largest source of theoretical uncertainty. The signal modelling is cross-checked between two independent generators, with the BlackMax/CHARYBDIS cross-section ratio quoted explicitly (1.36, 1.59, 1.78 for n = 2, 4, 6), and the limits are shown for non-rotating, rotating, and stable-remnant evaporation scenarios in Fig. 4, so the model dependence is displayed rather than asserted. The model-independent limits of Fig. 5 are a reusable asset for the community and are correctly advertised as such. I also checked the arithmetic of Table 1 against the quoted luminosity and acceptances (n_sig = sigma A L reproduces the listed values) and the crossing points of the Table 1 cross sections with the sigma_95 column against the mass limits quoted in the text and Fig. 3; they agree. The paper is internally consistent on the points where a reader is","major_comments":[{"comment":"The normalization procedure for the N ≥ 4 and N ≥ 5 predictions is stated ambiguously, and the ambiguity bears directly on the quoted systematic. The text says the solid curves are 'found by normalizing the fits of the N = 2 and 3 S_T distributions to the range S_T = 1000−1100 GeV', which does not say whether the N ≥ 4 and N ≥ 5 curves are normalized to their own data in that window or inherit the N = 2,3 normalization. This matters because the stated 4–12% normalization uncertainty, 'dominated by the statistics in the normalization region', is plausible for the high-statistics N = 2,3 samples but would be far larger if computed from the N ≥ 5 sample, where Fig. 2c shows of order one event per 100 GeV bin. Please state the procedure explicitly for each multiplicity and, if the normalization is per-multiplicity, quote the observed event counts in the 1000–1100 GeV window for N ≥ 3, 4, and 5 so the reader can verify the quoted normalization uncertainty at the working points that drive the 4.2 and 4.5 TeV limits.","section":"Background normalization paragraph preceding Figs. 2a–c","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The extrapolation of the S_T shape from N = 2,3 to N ≥ 4,5 is the least-tested link in the mass limit, and the 6–125% systematic band is derived from alternate ansatz functions plus the N = 2 versus N = 3 difference, i.e. a two-point check at adjacent multiplicities. The authors acknowledge the assumption and price it in, which is the right response, but the paper would be stronger if the size of the effect at the actual working points were shown rather than left implicit in a band. Concretely: for the N ≥ 5 working point (S_T^min = 2.9 TeV, Table 1), what is the predicted background under each of the three ansatz functions, and by how much does the 4.5 TeV limit move across the extremes of the band? A second, cleaner test is available and not used: the N ≥ 4 and N ≥ 5 samples have a signal-free region below S_T ≈ 1100 GeV (the turn-on is only N × 50 GeV), so the shape could be fitted directly there and compared with the N = 2,3 extrapolation. As it stands, the agreement visible in Fig. 2c above 1100 GeV is in the signal region and therefore cannot separate a shape error from a signal deficit.","section":"Multiplicity-independence of the S_T shape; Figs. 2b,c and Table 1","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The selection columns are printed as 'N ≥ N_min' and 'S_min^T'; please state in the caption that events are counted with N ≥ N_min and S_T > S_T^min, and repeat the definitional asymmetry that N counts only objects with E_T > 50 GeV while S_T also includes missing transverse energy when it exceeds 50 GeV.","section":"Table 1 caption and column headers","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The default ansatz is printed as P0(1+x)^P1 / x^P2+P3 log(x), whose grouping is ambiguous as typeset; please render it so that the exponent of x is unambiguously P2 + P3 log x, and give the fitted parameter values and a goodness-of-fit measure (chi-squared per degree of freedom) for the N = 2 and N = 3 fits so the reader can judge the ansatz independently of the two alternates.","section":"Fit ansatz paragraph","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The assumptions of no energy loss to gravitational shock waves and no graviton emission are stated but their direction of effect on the limits is not. Since both assumptions maximize the signal for a given M_BH^min, the resulting limits are conservative in that direction; saying so explicitly would help the reader interpret Fig. 4.","section":"Signal modelling paragraph","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The statement that the trigger is 'fully efficient for the offline analysis selections described below' should be quantified, for example as a measured plateau efficiency with its uncertainty, since the HLT jet thresholds varied between 20 and 30 GeV across the run.","section":"Trigger paragraph","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The abstract quotes 35 pb^-1 while the text quotes 34.7 ± 3.8 pb^-1; harmonize the two so the luminosity uncertainty is visible where the number first appears.","section":"Abstract versus data-set paragraph","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Reference [13] duplicates the collaboration name ('CMS Collaboration, CMS Collaboration') and its DOI appears truncated as 10.1103/PhysRevLett.105.21180 where the article number is 211801. Reference [12] lists J. Phys. G37 (2010) 105108 but a DOI ending 37/10/105008; please check the page number against the DOI.","section":"References [12] and [13]","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The axis labels of Fig. 5 appear garbled in the version I read ('b)rho A (x) minT > ST (S sigma'); please check them in proof. There is also a typographical error, 'uncertainy', in the paragraph assigning the 6% to 125% background systematic, and the date '2018/10/22' on the title page appears to be a template artifact.","section":"Fig. 5 axis labels and systematic paragraph","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision","confidential_remarks":"This is a sound, well-executed Letter and the physics conclusion is not in doubt on my reading; the numbers in Table 1 reproduce from the quoted luminosity and acceptances, and the mass limits are consistent with the cross-section limits in Fig. 3. My recommendation of minor revision is driven almost entirely by one presentational point: the normalization procedure for the N ≥ 4,5 background is described in a single ambiguous sentence, and since that procedure determines whether the quoted 4–12% normalization uncertainty is credible at the N ≥ 5 working point, it should be spelled out before publication. The multiplicity-extrapolation concern is acknowledged and priced in by the authors and does not, in my view, threaten the central claim."},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"This is the paper that made microscopic black holes an experimental subject rather than a seminar joke. CMS took 35 pb^-1 of 7 TeV data, selected high-S_T multi-object events, saw nothing above a data-derived QCD background, and set 95% CL lower limits of 3.5–4.5 TeV on the minimum black hole mass for M_D up to 3.5 TeV. The 'first direct limits' claim in [abstract:s4] holds up; no prior dedicated collider search is cited, and I don't know of one. The model-independent cross section times acceptance limits in Fig. 5 are the part with the longest shelf life — they outlive the specific ADD benchmarks and are usable against string balls, R-parity violating gluinos, and whatever else produces high-multiplicity high-S_T final states. That is the real contribution, and it is done properly: object definitions, trigger, fit ansatz with two alternates, normalization window, nuisance parameters and priors are all stated, and Table 1 gives signal, background, data, and both observed and expected limits at every working point. The paper is also unusually candid about model dependence — the BlackMax/CHARYBDIS cross section spread (1.36–1.78) and the M_D convention factor (up to 9.29 for n=6) are stated in the text rather than buried. That is good practice and I wish more searches did it. The soft spot is the one the stress-test names: the N>=4 and N>=5 backgrounds are never fit directly, only extrapolated from N=2 and N=3 fits under the S_T shape-invariance assumption, and the 6–125% systematic is the honest price of that. But I'd push back on how much weight the concern carries. The physical argument — that S_T in QCD is set by the hard 2->2 scattering and is therefore insensitive to how the event fragments — is sound, not just an empirical fit, and the observed N>=5 data in Fig. 2c sits on or below the predicted curve in the signal region. Since a signal can only add, a shape error making the background too low would have to be cancelled by a real deficit, which is not a serious possibility here. The concern is real but bounded by the data itself, and the authors have already paid for it in the systematic band. Minor points: the abstract rounds 34.7 to 35 pb^-1, and there is no comparison to indirect M_D bounds from astrophysics or precision tests, which would have been useful context. Neither is a problem. This deserves referees and should be published; it is the reference point for every black hole search that follows. I'd cite it.","headline":"A genuine first: a clean, honest, data-driven search that turns microscopic black holes from a thought experiment into a bounded scenario, with the main soft spot being the multiplicity extrapolation of the QCD background — which the authors price in and the data partly police.","tokens_in":30258,"tokens_out":1259,"would_cite":true,"duration_ms":22972,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":["04.70.Dy","11.10.Kk","13.85.-t"],"model":"selfhosted/deepseek-v4.1-flash","headline":"Using 35 inverse picobarns of 7 TeV proton collisions, CMS finds no excess of high-multiplicity, high-transverse-energy events and turns that null result into the first direct collider limits on microscopic black hole production, excluding ","keywords":["microscopic black holes","large extra dimensions","ADD model","CMS","LHC","multijet background","Hawking radiation","model-independent limits"],"falsifier":"A black hole signal would appear as an excess at high $S_T$ in the $N$≥4 or $N$≥5 distributions of Fig. 2; conversely, the background method would be falsified by data in a signal-free $S_T$ window at $N$≥4 whose shape differs from the $N=2,3$ fit beyond the quoted 6–125% systematic. Either outcome is decidable with more 7 TeV data, or with the same analysis repeated at 13 TeV.","tokens_in":30558,"feed_emoji":"🕳️","tokens_out":3743,"duration_ms":75567,"temperature":0.7,"pith_summary":"If gravity becomes strong at the TeV scale, as models with large flat extra dimensions propose, then protons colliding at 7 TeV should occasionally collapse into microscopic black holes that evaporate into a dozen or so energetic particles at once. The CMS collaboration looks for exactly that: events whose total transverse energy $S_T$ is spread over many jets, leptons, and photons, in 35 pb$^{-1}$ of 2010 data. Nothing stands out above the expected quantum-chromodynamic multijet background, so the paper converts the null result into 95% confidence upper limits on black hole production. The payoff is a first direct, accelerator-based exclusion of the low-mass end of that parameter space, together with a set of limits that do not depend on any particular black hole model.","feed_headline":"No black holes below 3.5 TeV in first LHC search","feed_subtitle":"35 inverse picobarns of 7 TeV collisions show no excess, excluding low-scale gravity black holes up to 4.5 TeV.","key_machinery":"The load-bearing device is the total transverse energy $S_T$, the scalar sum of $E_T$ over all selected objects above 50 GeV (plus missing transverse energy when it exceeds 50 GeV), combined with the object multiplicity $N$. The argument rests on a claimed shape invariance: because $S_T$ in QCD events is set almost entirely by the hard $2→2$ parton scattering, the shape of its distribution should not depend on how many jets, leptons, or photons end up reconstructed. That lets the collaboration fit the $S_T$ spectrum of the low-multiplicity $N=2$ and $N=3$ samples between 600 and 1100 GeV with an ansatz $P_0(1+x)^{P_1}/(x^{P_2}+P_3x^{P_3})$ — actually $P_0(1+x)^{P_1}/(x^{P_2}+P_3x^{P_3})$ in the paper's form $P_0(1+x)^{P_1}/(x^{P_2}+P_3x^{P_3})$ — and then normalize that fit in the signal-free window $S_T = 1000$–$1100$ GeV to predict the background at $N$≥4 and $N$≥5. The same shape is checked against two alternative fit functions, against Monte Carlo multijet generators, and against the mutual agreement of the $N=2$ and $N=3$ fits; the resulting background systematic spans 6% to 125% across the $S_T$ range used.","core_discovery":"The paper establishes that in $pp$ collisions at $7$ TeV, the observed $S_T$ spectra for events with $N=2$, $N=3$, $N=4$, and $N=5$ reconstructed objects agree with a background predicted from the data themselves, dominated by QCD multijet production. No excess appears, so the collaboration sets 95% confidence upper limits on the black hole production cross section as a function of the minimum black hole mass. For the three benchmark points $(M_D, n) = (1.5~TeV, 6)$, $(2.0~TeV, 4)$, $(3.0~TeV, 2)$, the observed lower limits on $M_{BH}^{min}$ are 3.5, 4.2, and 4.5 TeV (expected 3.2, 4.0, and 4.5 TeV). Translated into the ADD parameter space, this excludes black holes with minimum mass 3.5–4.5 TeV for multidimensional Planck scales up to 3.5 TeV at 95% CL, with similar reach for rotating black holes and for the stable-remnant scenario. The paper also publishes model-independent upper limits on cross section times acceptance for inclusive $N$≥3, 4, 5 final states above a threshold $S_T^{min}$. The authors describe these as the first direct limits on black hole production at a particle accelerator.","pith_inferences":["The shape-invariance premise is itself testable with the same data at higher integrated luminosity: once the $N$≥4 samples accumulate enough events in the 600–1100 GeV window, the background can be fitted directly there instead of extrapolated.","Since the geometric production cross section falls steeply with $M_{BH}^{min}$, the reach of this search is limited more by collected luminosity than by collision energy at the low-mass end, so a modest increase in dataset buys a modest increase in excluded mass rather than a step change.","The insensitivity of the limits to rotation and to remnant formation suggests that high multiplicity plus high $S_T$ is a fairly generic signature of any TeV-scale gravity that liberates many hard partons, which is why the model-independent limits travel further than the specific ADD benchmarks.","Quantum black holes that decay promptly to two jets are invisible to this analysis by construction — the paper explicitly defers them to dedicated dijet searches — so a joint multiplicity-plus-dijet combination would close a gap that neither analysis covers alone."],"forward_implications":["ADD models with multidimensional Planck scale up to 3.5 TeV and minimum black hole mass between 3.5 and 4.5 TeV are excluded at 95% confidence, closing the region where black hole production at the LHC would have been copious.","Because rotating black holes, non-rotating black holes, and the stable non-interacting remnant scenario all give similar exclusions, the limit is not an artifact of one particular evaporation model.","The model-independent $\\sigma \\times A$ limits can be reused to constrain any new physics that yields energetic, high-multiplicity final states, including high-mass $t\\bar{t}$ resonances, R-parity-violating gluino decays into three jets, and string balls.","The analysis defines the template — data-driven multijet background plus $S_T$-and-multiplicity counting — that later searches at higher energy and luminosity inherit."],"supporting_citations":[{"why":"Supplies the original black hole production cross section from geometric considerations and the alternative definition of the Planck scale used for comparison.","marker":"[1]"},{"why":"Independent derivation of the same production picture, used jointly with [1] for the cross section and Hawking temperature.","marker":"[2]"},{"why":"Defines the large flat extra dimension model whose parameters $M_D$ and $n$ the limits are quoted in.","marker":"[3]"},{"why":"Supplies the thermal Hawking evaporation that makes the black hole decay into the many-object final state being searched for.","marker":"[8]"},{"why":"Defines the anti-$k_T$ jet clustering algorithm used to reconstruct the objects entering $S_T$ and $N$.","marker":"[19]"},{"why":"The BlackMax event generator that produces the signal samples and hence the acceptances and predicted event yields in Table 1.","marker":"[23]"},{"why":"CHARYBDIS generator used to cross-check cross sections and to simulate the stable non-interacting remnant scenario that BlackMax cannot produce.","marker":"[28]"},{"why":"The Bayesian limit recipe with log-normal nuisance priors used to extract the 95% confidence upper limits on cross section.","marker":"[33]"}],"fun_headline_variants":["CMS finds no black hole excess in 7 TeV collisions","First direct limits on microscopic black holes from CMS","No black hole signal in 35 inverse picobarns at 7 TeV","CMS excludes low-scale gravity black holes up to 4.5 TeV"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The background expected in four- and five-object events is obtained by fitting the two- and three-object spectra and extrapolating that shape to higher multiplicity, which assumes the total-transverse-energy distribution does not change with the number of reconstructed objects — an assumption the paper checks with generators and with the agreement of its own $N=2$ and $N=3$ fits, but never with signal-free data at $N$≥4, and which is why it assigns a background systematic as large as 125%.","fun_headline_variants_meta":{"raw":{"variants":["CMS finds no black hole excess in 7 TeV collisions","First direct limits on microscopic black holes from CMS","No black hole signal in 35 inverse picobarns at 7 TeV","CMS excludes low-scale gravity black holes up to 4.5 TeV"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":1341,"prompt_tokens":1002,"completion_tokens":339,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":275}},"tokens_in":1002,"tokens_out":339,"duration_ms":6722,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":275,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-11-10T18:26:31.152412+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"A black hole signal would appear as an excess at high $S_T$ in the $N$≥4 or $N$≥5 distributions of Fig. 2; conversely, the background method would be falsified by data in a signal-free $S_T$ window at $N$≥4 whose shape differs from the $N=2,3$ fit beyond the quoted 6–125% systematic. Either outcome is decidable with more 7 TeV data, or with the same analysis repeated at 13 TeV.","supporting_citations":[{"cited_title":"Black Holes at the LHC","cited_arxiv_id":"hep-ph/0106295","evidence_quote":"Supplies the original black hole production cross section from geometric considerations and the alternative definition of the Planck scale used for comparison."},{"cited_title":"High Energy Colliders as Black Hole Factories: The End of Short Distance Physics","cited_arxiv_id":"hep-ph/0106219","evidence_quote":"Independent derivation of the same production picture, used jointly with [1] for the cross section and Hawking temperature."},{"cited_title":"The Hierarchy problem and new dimensions at a millimeter","cited_arxiv_id":null,"evidence_quote":"Defines the large flat extra dimension model whose parameters $M_D$ and $n$ the limits are quoted in."},{"cited_title":"Particle creation by black holes","cited_arxiv_id":null,"evidence_quote":"Supplies the thermal Hawking evaporation that makes the black hole decay into the many-object final state being searched for."},{"cited_title":"The anti- <i>k</i> <sub> <i>t</i> </sub> jet clustering algorithm","cited_arxiv_id":null,"evidence_quote":"Defines the anti-$k_T$ jet clustering algorithm used to reconstruct the objects entering $S_T$ and $N$."},{"cited_title":"BlackMax: A black-hole event generator with rotation, recoil, split branes, and brane tension","cited_arxiv_id":null,"evidence_quote":"The BlackMax event generator that produces the signal samples and hence the acceptances and predicted event yields in Table 1."},{"cited_title":"CHARYBDIS: A Black Hole Event Generator","cited_arxiv_id":"hep-ph/0307305","evidence_quote":"CHARYBDIS generator used to cross-check cross sections and to simulate the stable non-interacting remnant scenario that BlackMax cannot produce."},{"cited_title":"ALPGEN, a generator for hard multiparton processes in hadronic collisions","cited_arxiv_id":"hep-ph/0206293","evidence_quote":"The Bayesian limit recipe with log-normal nuisance priors used to extract the 95% confidence upper limits on cross section."}],"review_version":1}