{"id":"500bce07-ba65-4c5c-84af-41f35612e5bb","arxiv_id":"2412.14247","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Selecting close, asymmetric dijets at the LHC gives a ~90% gluon-enriched subleading jet sample, supported by fixed-order QCD and Monte Carlo.","lead":"A collinear, momentum-imbalanced dijet selection makes the subleading jet act like a secondary Lund plane and yields about 90% gluon-initiated jets, a useful LHC handle for gluon-jet measurements.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Hadron-level MC validates the density proxy but never directly measures the claimed 90% gluon fraction; the headline purity rests on LO parton-level estimates.","rationale":"The reader's weakest assumption was that the subleading anti-kt jet is identified with the soft branch of a primary splitting, i.e., the secondary Lund plane, relying on the MC comparison in Sec. 3. My concern is adjacent but more specific: even granting that mapping, the MC comparison is a density-level check, not a direct purity measurement. The density agreement within 10% with the gg->gg primary Lund plane is strong indirect evidence, but the abstract's quantitative claim is 'gluon jet fractions of around 90%... confirmed using hadron-level Monte Carlo.' Since the paper is otherwise careful about the ambiguity of quark/gluon labels, the missing direct truth-level fraction is a clear, fillable gap rather than a fatal flaw. This does not change the reader's CONDITIONAL verdict: the paper remains a promising proposal conditional on a direct hadron-level purity extraction and, ideally, public implementation. I therefore recommend no change to the verdict, while flagging the direct purity check as the load-bearing validation step that should be added before the headline number is relied upon.","tokens_in":18742,"tokens_out":11304,"duration_ms":105711,"concrete_test":"In the same Pythia8 and Herwig7 hadron-level samples used for Figs. 3-6, tag the subleading jet (leading pT>700 GeV, subleading 150<pT<200 GeV, 1<DeltaR<1.2, |y|<1.7) using the MC truth record: define the jet-initiating flavor by the highest-pT parton inside the jet after the hard process, and also test an alternative definition (the final-state parton with the largest momentum fraction along the jet axis). Compute the gluon fraction for all QCD dijet events and separately for qq->qq and gg->gg hard scatterings, as a function of DeltaR and pT,sub. If the fraction is 90% within a few percent in both channels, the claim is confirmed; if it falls below about 80% in either channel, the headline purity should be revised or the interpretation as a gluon-enriched sample should be weakened.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central quantitative claim is that the subleading jet in the dijet selection (pT,lead>700 GeV, 150<pT,sub<200 GeV, 1<DeltaR<1.2) is gluon-initiated about 90% of the time, for both quark- and gluon-initiated hard scatterings. The analytic support (Sec. 2.2) is a leading-order, double-logarithmic calculation of the probability that the soft branch of a single splitting is a gluon. The hadron-level MC studies in Sec. 3, however, never report a truth-level gluon fraction for the selected subleading jets. Instead, they compare the secondary Lund plane density to the primary Lund plane density of jets from gg->gg scatterings (Figs. 4-6) and show agreement within 10%. That density agreement is indirect evidence: it constrains the color factor of the emissions, but it does not by itself establish that 90% of the subleading jets are gluon-initiated, because the density is an average over all emissions and all jet flavors. Given the paper's own operational definition of gluon-enrichment as an MC-interpreted quantity (footnote 2), a direct truth-label purity extraction is the missing piece that would convert the LO estimate into a hadron-level statement. Without it, the abstract's claim that the 90% observation is 'confirmed using hadron-level Monte Carlo generated events' is not fully supported by the presented evidence.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a new experimental strategy to obtain a gluon-enriched jet sample at the LHC by selecting a dijet pair with highly asymmetric transverse momenta and small angular separation (nominal cuts: pT,lead > 700 GeV, 150 < pT,sub < 200 GeV, 1 < DeltaR < 1.2) and measuring the primary Lund plane density of the subleading jet. The authors argue that this subleading jet is practically equivalent to the soft branch of a primary splitting, i.e., a secondary Lund jet plane, and hence is dominantly gluon-initiated. They support this with a leading-order fixed-order calculation (collinear Altarelli-Parisi estimates and exact O(alpha_s) NLOJet++ results) giving gluon fractions around 90% for both quark- and gluon-initiated hard scatterings, and with hadron-level Monte Carlo studies using Pythia8 and Herwig7 that compare the subleading-jet Lund plane density with the primary Lund plane density of gluon jets from gg->gg scatterings, finding agreement within 10%. The paper also studies dependence on the hard-scattering flavor, g->qqbar splittings, and MC generator choices, and discusses applications to MC tuning and alpha_s extraction.","tokens_in":19089,"tokens_out":5243,"duration_ms":45338,"significance":"If the central claim holds, the proposal offers a simple, experimentally accessible way to obtain high-purity gluon jets without taggers or statistical demixing, which would be valuable for constraining gluon fragmentation in MC generators and for reducing the quark/gluon-fraction degeneracy in alpha_s extractions. The paper's strengths include a parameter-free analytic estimate using standard splitting functions, a cross-check with exact fixed-order matrix elements, and a broad MC validation with two generators and three parton-shower variants. The fixed-order derivation is clean and the density comparisons are suggestive. However, the headline 90% purity is supported only at leading order, and the hadron-level MC confirmation is indirect because it never measures the truth-level gluon fraction of the selected jets. The paper is a useful phenomenological proposal, but the strength of the claims needs to be tempered or the missing direct validation added.","major_comments":[{"comment":"The abstract and Sec. 4 state that the ~90% gluon fraction is 'confirmed using hadron-level Monte Carlo generated events,' but Sec. 3 never reports a truth-level gluon purity for the selected subleading jets. The comparisons in Figs. 3-6 are between Lund plane densities (the secondary density of the subleading jet and the primary density of gluon jets from gg->gg scatterings), which constrain the average color factor of emissions but do not directly measure the fraction of gluon-initiated jets. Because the paper itself adopts an MC-interpreted definition of gluon-enrichment (footnote 2), a direct extraction of the truth-label gluon fraction should be provided, or the abstract and conclusions should be reworded to state that the MC supports the density proxy rather than confirming the 90% purity.","section":"Sec. 3"},{"comment":"The dijet selection identifies the subleading anti-kT jet with the soft branch of a primary C/A splitting without derivation. The analytic estimate in Eq. (2.9) uses the splitting variable z = pT,sub/(pT,lead+pT,sub) and the angular separation DeltaR as the splitting angle, which is only valid if the two jets are the two branches of a single primary emission. In the nominal dijet setup, the two jets are separate R=0.4 anti-kT jets, and the subleading jet is not constructed by declustering a common parent. The MC comparison in Sec. 3 is the only evidence for this equivalence; it shows density agreement but not that the per-jet identification holds. The 90% purity should be either verified by matching the subleading jet to the soft branch of the C/A tree in MC, or explicitly stated as applying to the idealized single-splitting configuration.","section":"Sec. 2.1/2.2"},{"comment":"The purity estimate is leading-order and double-logarithmic, with resummation and higher-order corrections postponed (as acknowledged in the text). While this is an acceptable first step for a proposal, the central quantitative claim of 'around 90%' is presented without an estimate of perturbative uncertainty; the exact O(alpha_s) NLOJet++ results in Fig. 2 are close to the collinear approximation, but both are leading order in the relevant splitting. To support the headline number, the authors should either provide an estimate of higher-order effects (e.g., a simple resummed estimate or a scale-variation band on the purity) or clearly qualify the 90% as a leading-order estimate in the abstract.","section":"Sec. 2.2"}],"minor_comments":[{"comment":"The word 'infrarred' in 'infrarred-and-collinear (IRC) safe' should be corrected to 'infrared-and-collinear'.","section":"Sec. 2"},{"comment":"The lower panels are labeled 'Ratio-to-incl.' but the captions do not state explicitly whether the bands are statistical only; please clarify in the caption.","section":"Fig. 3"},{"comment":"The statement that about O(10^5) 'gluon-like jets' are obtained should define how 'gluon-like' is determined at generator level (e.g., by the hard-scattering process or by parton flavor), to avoid ambiguity.","section":"Sec. 3"},{"comment":"Eq. (2.8) defines the gluon fraction conditional on a leading quark or gluon, but the text does not give the formula for the inclusive purity that combines both; adding this would make the transition from Fig. 2 to the inclusive MC results clearer.","section":"Sec. 2.2"}],"recommendation":"major_revision","confidential_remarks":"The paper fits the scope of JHEP and the fixed-order calculation is sound. The main concern is that the abstract overclaims the hadron-level confirmation: the MC studies validate the Lund plane density proxy but do not directly measure the 90% gluon purity. This is a fixable issue: the authors should add a truth-level purity extraction in MC or qualify the claim. A focused revision along the lines of the major comments is appropriate; no concerns about novelty or citation pattern."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Good paper, worth engaging with. The core idea is simple and new: use a dijet selection with a large momentum imbalance and small angular separation, and treat the subleading jet as a proxy for the secondary Lund plane. The soft-singularity argument behind it is old, but turning it into a concrete, tagger-free LHC selection with ~90% gluon purity is a genuine contribution. The fixed-order calculation is clean, uses the standard Altarelli-Parisi kernels with no fitted parameters, and reproduces the observed trends. The hadron-level MC work is solid: two generators, hadron-level final state, UE and MPI included, and the comparison to a gg->gg primary Lund plane is a sensible consistency check. The claimed insensitivity to the hard-scattering flavor and PDFs is a useful feature.\n\nThe soft spots are real but not fatal. The 90% purity is an LO double-log estimate; the paper says resummation is postponed, which is honest. The bigger issue is that the MC never directly reports the gluon fraction of the selected subleading jets. The density agreement within 10% is indirect evidence; it constrains the color factor of emissions but does not literally confirm the 90% number. The abstract's wording, 'confirmed using hadron-level Monte Carlo generated events', overstates what the MC actually shows. A truth-level purity extraction would close this gap cleanly. There is also an assumption, acknowledged in Sec. 2.1, that the subleading anti-kt jet is 'practically equivalent' to the secondary Lund plane; the MC comparison gives some confidence, but a direct check of how often the subleading jet corresponds to the soft branch of the primary splitting would strengthen the paper.\n\nOverall, this is a credible proof-of-concept. It should be sent to peer review. I would ask the authors to add a direct MC purity measurement, or soften the language in the abstract. The paper is useful for jet substructure practitioners, MC tuners, and LHC experimentalists looking for a simple gluon-enriched sample.","headline":"A simple, credible new route to gluon-enriched jet samples; the 90% purity rests on solid LO work and an indirect but decent MC consistency check.","tokens_in":19598,"tokens_out":2681,"would_cite":true,"duration_ms":23933,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The subleading jet of a collinear, asymmetric dijet pair is gluon-initiated about 90% of the time, making its primary Lund plane density a practical proxy for the gluon-rich secondary Lund plane.","keywords":["gluon jets","Lund jet plane","secondary Lund plane","dijet selection","jet substructure","quark-gluon discrimination","LHC","parton shower tuning"],"falsifier":"Measure the parton flavour of the subleading jet with a quark/gluon discriminator on real LHC data in exactly the fiducial region $p_{t,\\mathrm{lead}}>700$ GeV, $150<p_{t,\\mathrm{sublead}}<200$ GeV, and $1<\\Delta R<1.2$: if the gluon fraction comes out well below 90%, or if the measured subleading-jet Lund plane density deviates from the primary Lund plane of $gg\\to gg$ gluon jets by more than the claimed 10% in any $(\\ln k_t,\\ln 1/\\Delta)$ bin, the equivalence breaks.","tokens_in":18539,"feed_emoji":"⚛️","tokens_out":5537,"duration_ms":47112,"temperature":0.7,"pith_summary":"The paper proposes a phase-space selection at the LHC that produces a gluon-enriched sample of jets without taggers or statistical demixing: take a dijet pair with a hard leading jet ($p_t>700$ GeV), a much softer subleading jet ($150<p_t<200$ GeV), and a small angular separation ($1<\\Delta R<1.2$). Because the soft divergence of QCD splitting functions makes the softer branch of a hard splitting preferentially a gluon, the subleading jet is gluon-initiated about 90% of the time for both quark- and gluon-initiated hard scatterings. Measuring the primary Lund plane density of that subleading jet therefore acts as a measurement of the secondary Lund plane, the radiation pattern of the soft gluon branch. Fixed-order calculations give the ~90% purity, and hadron-level Monte Carlo confirms that the density matches the primary Lund plane of pure gluon jets within 10%. The strategy matters because gluon-jet fragmentation is among the least constrained ingredients in LHC Monte Carlo tuning and jet energy calibration.","feed_headline":"Dijet selection yields ~90% gluon jets at the LHC","feed_subtitle":"A collinear, asymmetric dijet pair turns the subleading jet into a gluon-rich Lund plane without taggers.","key_machinery":"The central object is the secondary Lund plane density $\\rho_s$: the density of emissions obtained by recursively declustering the softer subjet of the primary Cambridge/Aachen declustering, plotted in $(\\ln k_t,\\ln 1/\\Delta)$. Its power comes from the QCD soft singularity, which makes the soft branch of an asymmetric splitting usually a gluon, so $\\rho_s$ is dominated by colour factor $C_A$. The dijet selection turns this abstract plane into a measurable object: the subleading jet of a collinear, asymmetric pair is treated as the secondary plane, so its primary Lund plane density, computed with the standard anti-$k_t$ reclustering, is the observable.","core_discovery":"The central claim is that the secondary Lund plane, the set of emissions off the softer branch of the primary splitting, is a gluon-dominated object, and that a simple dijet selection makes it experimentally accessible. At leading order the secondary emission's colour factor is $C_A$ whenever the split is $q\\to qg$ with a soft gluon or $g\\to gg$, while the $g\\to q\\bar q$ channel is suppressed because it lacks a soft singularity. With $p_{t,\\mathrm{lead}}>700$ GeV, $150<p_{t,\\mathrm{sublead}}<200$ GeV, and $1<\\Delta R<1.2$, the subleading anti-$k_t$ jet is \"practically equivalent\" to the secondary Lund plane and is gluon-initiated in about 90% of cases, for both quark- and gluon-initiated events. The gluon purity is resilient to the colour structure of the event, the hard-scattering flavour, and PDF choice, and the resulting secondary Lund plane density agrees with the primary Lund plane density of gluon jets from $gg\\to gg$ within 10%.","pith_inferences":["Inference: If the mapping between the subleading jet and the secondary Lund plane survives experimental scrutiny, the same dijet selection could provide gluon-enriched samples for many other substructure observables, such as hadron chemistry, jet charge, and energy correlators, not just Lund plane densities.","Inference: The same collinear-and-asymmetric dijet selection could be adapted to heavy-ion collisions as a way to isolate the colour-charge dependence of jet energy loss, though underlying-event contamination would need to be controlled first.","Inference: A direct experimental check of the claimed purity would be to apply existing quark/gluon discriminators to the selected subleading jets and compare the extracted gluon fraction with the ~90% prediction, a test the paper itself suggests.","Inference: Extending the selection to a trijet configuration with two b-tagged jets and an anti-b tagged third jet could offer a gluon-enriched sample with additional heavy-flavour handles, as the paper notes in passing."],"forward_implications":["An LHC measurement of the primary Lund plane density of the subleading jet can constrain gluon-initiated radiation without relying on quark/gluon taggers or statistical demixing.","The approximately 90% gluon purity holds for both quark- and gluon-initiated hard scatterings, making the sample largely insensitive to PDF choice and to the flavour composition of the hard process.","The secondary Lund plane density from inclusive dijets reproduces the primary Lund plane density of $gg\\to gg$ gluon jets within 10%, so it can serve as a practical proxy for pure gluon jets.","Differences between Monte Carlo generators in the deep collinear region appear in the secondary plane just as in the pure gluon primary plane, giving the observable similar constraining power for shower and hadronisation models.","The selection uses standard anti-$k_t$ jets with $R=0.4$ and a simple fiducial region, so it can be applied directly to existing LHC data and implemented in routine Rivet analyses.","Higher gluon purities can be reached by selecting softer subleading jets, at the cost of reduced phase space."],"supporting_citations":[{"why":"Defines the Lund jet plane and the recursive declustering procedure that the paper uses to construct the secondary Lund plane.","marker":"[53]"},{"why":"Provides the resummation framework for the primary Lund plane density that the paper reuses for the secondary Lund plane density.","marker":"[58]"},{"why":"Defines the anti-$k_t$ jet algorithm used for the dijet selection and for the jets whose substructure is measured.","marker":"[64]"},{"why":"Supplies the exact $O(\\alpha_s)$ three-jet calculation with flavour information used to estimate gluon purities at fixed order.","marker":"[68]"},{"why":"Supplies the PDF set used in both the fixed-order and Monte Carlo estimates, which the paper shows to have negligible impact on the extracted purity.","marker":"[69]"},{"why":"Provides one of the hadron-level Monte Carlo generators used to confirm the gluon purity and the 10% agreement with the pure-gluon primary Lund plane.","marker":"[70]"},{"why":"Provides the other hadron-level Monte Carlo generator, with both angular-ordered and dipole showers, used for confirmation and model-comparison studies.","marker":"[71]"}],"fun_headline_variants":["Secondary Lund plane: 90% gluon purity without taggers","Collinear dijets expose gluon jets at 90% purity","Gluon-rich jets from secondary Lund plane at LHC","Simple dijet cut yields 90% gluon jet sample","Lund trick: isolate gluon jets to 90% purity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire scheme assumes that the subleading jet of the selected dijet pair really is the soft branch of one primary splitting; if ordinary jet clustering, underlying-event activity, or boundary effects break that correspondence, the predicted ~90% gluon purity would not apply to the measured jet.","fun_headline_variants_meta":{"raw":{"variants":["Secondary Lund plane: 90% gluon purity without taggers","Collinear dijets expose gluon jets at 90% purity","Gluon-rich jets from secondary Lund plane at LHC","Simple dijet cut yields 90% gluon jet sample","Lund trick: isolate gluon jets to 90% purity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000182,"raw_usage":{"total_tokens":1328,"prompt_tokens":978,"completion_tokens":350,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":594,"completion_tokens_details":{"reasoning_tokens":261}},"tokens_in":594,"tokens_out":350,"duration_ms":3829,"temperature":1.0,"reasoning_tokens":261,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T12:23:26.927836+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the parton flavour of the subleading jet with a quark/gluon discriminator on real LHC data in exactly the fiducial region $p_{t,\\mathrm{lead}}>700$ GeV, $150<p_{t,\\mathrm{sublead}}<200$ GeV, and $1<\\Delta R<1.2$: if the gluon fraction comes out well below 90%, or if the measured subleading-jet Lund plane density deviates from the primary Lund plane of $gg\\to gg$ gluon jets by more than the claimed 10% in any $(\\ln k_t,\\ln 1/\\Delta)$ bin, the equivalence breaks.","supporting_citations":[{"cited_title":"Measurements of Lund subjet multiplicities in 13 TeV proton-proton collisions with the ATLAS detector","cited_arxiv_id":"2402.13052","evidence_quote":"Defines the anti-$k_t$ jet algorithm used for the dijet selection and for the jets whose substructure is measured."}],"review_version":1}