{"id":"60b96f84-2025-44a7-bc61-a365f5ce44e7","arxiv_id":"1908.07524","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A top-partner search gap exists: for T -> t a with a -> gg, current LHC bounds drop from about 1.3 TeV to roughly 400-550 GeV.","lead":"Pair-produced heavy top-partner particles that decay through an intermediate light particle can evade current LHC searches, leaving masses as low as 400 GeV unexcluded. The paper maps this blind spot and tells experimenters where to look for the missing decay signature.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The mT≈400 GeV gap point is vulnerable: the excited-top recast assumes a from a T at rest, but pair-production boosts collimate a->gg, so the single-jet merging criterion in Eq.(A2) is underestimated and the grey bound could extend to ma≈50 GeV.","rationale":"The reader identified the excited-top merging assumption as the weakest point; I agree and sharpen it. The real mechanism is not merely 'detectors may merge differently' but that the paper's T-at-rest estimate of Δθ understates the collimation of a→gg for the boosted events that dominate the lepton+jets excited-top search. Because the central quantitative claim includes mT≈400 GeV at ma≈50 GeV, and because that point falls outside the paper's grey region only by virtue of the T-at-rest approximation, the exact value of the lower bound is not robust. The recast validations in Tables I–III and the explicit statement that the VLQ recast underestimates the bound are genuine independent checks, and they keep the broader 'search gap' claim credible. However, the excited-top recast has no validation and is not conservative in a well-defined direction: the T-at-rest criterion can either under- or over-estimate merging depending on the event boost. This does not invalidate the paper, but it means the 400 GeV number should be treated as conditional on an event-level check of the merging/acceptance assumption. The current CONDITIONAL verdict already captures this, so no change in verdict is needed.","tokens_in":15664,"tokens_out":21930,"duration_ms":818683,"concrete_test":"Generate pp→TTbar with mT=400 GeV, ma=50 GeV, BR(a→gg)=1 using MadGraph5+PYTHIA8+Delphes/FastJet, cluster with anti-kT R=0.4, and compute the fraction of events in which the two a partons are merged into a single jet (or have ΔR<0.4) after showering, together with the excited-top lepton+jets acceptance. Multiply by the Top++ pair-production cross-section and compare with the CMS 95% CL upper limit from Ref. [46]; if the expected signal exceeds the observed limit, the mT=400 GeV point is excluded and the headline lower bound must be raised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim's lowest point (mT≈400 GeV at ma≈50 GeV, BR(a→gg)=100%) is not protected by the searches the authors emphasize. For a T at rest, Eq.(A2) gives Δθ≈0.61 rad, far above the assumed 0.2 rad single-jet threshold, so the paper places this point outside the CMS excited-top grey region. But this is a lower bound on the boost: in QCD pair production the T is not at rest, and any boost of the T increases p_a and therefore collimates the a→gg pair. The excited-top search [46] is lepton+jets and preferentially selects boosted events, exactly the events where the T-at-rest estimate fails. For example, a T with pT≈600 GeV at mT=400 GeV boosts p_a to roughly 530 GeV, giving Δθ≈0.19<0.2; such events populate the high-pT tail of the pair-production distribution. The recast then multiplies the whole CMS upper limit by BR^2(T→ta) with no event-by-event merging fraction. If the tail acceptance is non-negligible, the grey exclusion extends toward ma≈50 GeV and the mT≈400 GeV point could be excluded. The paper itself states that the search is not cut-flow based and a recast is not possible, so the assumed equal-efficiency/merged-jet treatment is exactly the unvalidated step that supports the lowest allowed mass. This is a recast systematic, not an internal inconsistency, and it mainly attacks the 400 GeV point; the broader search-gap conclusion would survive with a somewhat higher lower bound.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies vector-like top partner T pair production in a simplified model where T can decay exotically as T -> t a, with a a light pseudoscalar decaying predominantly to gg or b bbar below the t tbar threshold. The authors recast three existing LHC searches (CMS excited-top pair production, ATLAS 8 TeV RPV SUSY multi-jet, and ATLAS 13 TeV up-type vector-like quark) into the (mT, ma) plane and into the BR(T -> th) vs mT plane. Their central result is that for BR(a -> gg) = 100%, the lower bound on mT is drastically weakened from the standard 1.3-1.4 TeV to about 400-550 GeV, leaving a 'search gap' at low ma. The paper includes cut-flow validation tables for the recasts and explicitly notes where the recasts underestimate the true bounds.","tokens_in":16022,"tokens_out":10620,"duration_ms":103310,"significance":"If the result holds, it identifies a well-motivated composite-Higgs decay chain that escapes the standard top-partner searches, which is a significant and timely observation. The paper is commendably transparent: the recasts are documented, validated against published event numbers in Tables I-III, and the authors flag where their treatment is approximate or conservative. The exclusion estimates are computed against independent ATLAS and CMS data rather than fitted to a desired conclusion. The main caveat, discussed below, is that the lowest-mass part of the claimed gap rests on an unvalidated assumption about jet merging and efficiency equivalence with excited-top production.","major_comments":[{"comment":"The single-jet merging criterion Delta theta < 0.2 is evaluated with the T at rest. In QCD pair production the T is typically boosted, and the lepton+jets selection of the excited-top search [46] preferentially selects boosted events; for these events the lab-frame separation of the a->gg pair is smaller than Eq. (A2) gives. For example, at mT = 400 GeV and ma = 50 GeV, Eq. (A2) gives Delta theta ~ 0.6 rad, so the paper places this point outside the grey region; but a T with pT ~ 600 GeV reduces Delta theta below 0.2 and such events populate the high-pT tail of the pair-production distribution. The grey excluded region in Fig. 2 is therefore likely underestimated, and the headline statement that mT ~ 400 GeV remains allowed for BR(a->gg)=100% is not protected by this recast. The merging criterion should be evaluated with the full T pT distribution, or a quantitative statement of why the at-rest approximation is conservative should be provided.","section":"Appendix A1, Eq. (A2) and Fig. 2"},{"comment":"The assumption that the signal efficiency for T and for an excited top quark of the same mass are equal is unvalidated. A merged a->gg jet has an invariant mass ma and a two-prong structure, whereas the gluon jet in excited-top production is a single-prong object, and the trigger and selection of [46] may respond differently to the two topologies. Because the authors state that a cut-flow recast of [46] is not possible, the grey region in Fig. 2 should be presented with an explicit systematic band, or with a robustness test such as varying the Delta theta threshold between 0.15 and 0.4, to show how much the boundary and the mT ~ 400 GeV point move.","section":"Appendix A1, single-jet efficiency assumption"},{"comment":"The conclusion that for larger ma the bound generically never passes ~550 GeV rests on the recast of the 8 TeV RPV search [47], while the 13 TeV RPV search [48] is dismissed as 'less sensitive' because of the high M_Sigma_J cut. Fig. 5 shows that a 1 TeV T pair has little support above the M_Sigma_J threshold, but it does not quantify the expected 13 TeV limit; for lower mT the T pair cross-section is much larger, and a small residual acceptance after the M_Sigma_J cut could produce a bound comparable to or stronger than the 8 TeV one. Please provide a quantitative estimate of the 13 TeV search's expected limit on the T->t a->t gg signal, or explicitly state that this is beyond the scope of the present recast and adjust the wording of the claim accordingly.","section":"Section II and Appendix A3, Fig. 5"}],"minor_comments":[{"comment":"The text reads 'Top parters'; this should be 'top partners'.","section":"Section I, first paragraph"},{"comment":"The wording 'mT ~ 400 GeV is still allowed' is internally inconsistent with the abstract's 'allowing a top partner mass as light as 400 GeV'; please harmonize the phrasing to clarify that the lower bound is around 400 GeV.","section":"Section II, paragraph after Fig. 2"},{"comment":"The black and grey solid lines are not separately labeled in the legend; please state explicitly in the caption which curve corresponds to the inclusive rescaling and which to the exclusive rescaling.","section":"Fig. 3 caption and legend"},{"comment":"The collaboration name appears as 'A TLASCollaboration'; fix the spacing.","section":"Bibliography, Ref. [6]"},{"comment":"The approximation symbol after the exact expression is unnecessary; also state the range of parameter values for which the approximate form, Delta theta ~ 4 ma mT / sqrt(...), is numerically accurate.","section":"Appendix A1, Eq. (A2)"}],"recommendation":"major_revision","confidential_remarks":"The paper is transparent, well documented, and addresses a timely question. The main risk is that the headline 'as light as 400 GeV' point depends on an unvalidated excited-top recast and an at-rest estimate that likely underestimates how often the a->gg pair is merged into a single jet. I recommend major revision rather than rejection: the authors can address the issue with a more realistic treatment of the T boost, or with an explicit scan over the merging threshold and an uncertainty band for the grey region. The broader message that exotic T->t a decays weaken the standard bound is likely robust, but the exact low-mass boundary needs work."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis is a solid, useful recast paper. It extends earlier exotic top-partner studies to charge-2/3 T -> t a with a -> gg/bb, and produces a mT-ma exclusion map from three LHC searches. The main finding — that for dominant a->gg decays the pair-production bound can drop to ~550 GeV, and in the low-ma window to ~400 GeV — is new relative to the cited searches and prior work, and the paper is honest about places where its recast underestimates the true bound (e.g., the VLQ search). The validation tables against ATLAS/CMS cut flows are credible, and the authors note the non-recastable nature of the excited-top search rather than hiding it.\n\nThe soft spots are real but addressable. The low-ma gg conclusion leans on Appendix A's assumption that a is tagged as a single jet when Δθ<0.2, computed from a T at rest, plus an equal-efficiency assumption relative to excited top pair production. The stress-test note is right that this is fragile: pair-produced T's are not at rest, and a boosted T of mT=400 with pT~600 can emit an a with p_a~530 GeV, giving Δθ~0.2 and making the a merge into a single jet. The excited-top search is lepton+jets and preferentially selects boosted events, so the grey excluded region could extend toward ma~50 GeV and exclude the mT~400 GeV point. The authors should either simulate the merging fraction or show that the high-pT tail is negligible after search selections. That said, this mainly attacks the lowest point on the exclusion curve; the broader statement that motivated exotic decays weaken the standard 1.3-1.4 TeV bound would survive with a somewhat higher lower bound.\n\nThere are also minor unquantified recast systematics, no public code, and the 8 TeV RPV recast gives a weaker bound than Ref[20], though the authors explain why.\n\nWho this is for: anyone working on vector-like quark searches or composite Higgs phenomenology. It deserves a serious referee — the central claim is interesting and supported enough to justify a careful check of the jet-merging assumption, not a desk reject. I would accept it for peer review, and I'd cite it if I worked in this area.\n\nRecommendation: engage with it; ask the authors to quantify the single-jet merging efficiency and the boost effect before acceptance.","headline":"A careful recast that likely identifies a real gap in top-partner coverage, though the 400 GeV point rests on a fragile single-jet assumption that could shift the lower bound upward.","tokens_in":16570,"tokens_out":4927,"would_cite":true,"duration_ms":52298,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["12.60.Rc","14.65.Ha","13.85.Rm","12.60.-i"],"model":"deepseek-v4-flash","headline":"A charge-2/3 top partner decaying through a light pseudo-scalar to gluons can evade current LHC searches, leaving a mass bound near 400 GeV instead of 1.3–1.4 TeV.","keywords":["vector-like top partner","composite Higgs","exotic decays","pseudo-scalar","LHC searches","recast","dijet resonance","top partner mass bound"],"falsifier":"Take $m_a=50$ GeV and $\\mathrm{BR}(a\\to gg)=1$, and ask whether the CMS 13 TeV excited-top dataset excludes $T\\bar T$ with $m_T=400$ GeV once the $a\\to gg$ system is simulated with full shower and detector response rather than the $\\Delta\\theta<0.2$ single-jet assumption. A 95% CL bound excluding $m_T=400$ GeV would falsify the paper's headline gap; an observed signal would confirm it.","tokens_in":15482,"feed_emoji":"⚛️","tokens_out":7021,"duration_ms":67935,"temperature":0.7,"pith_summary":"For a charge-2/3 vector-like top partner $T$ in composite Goldstone Higgs models, the standard LHC lower bound of about $1.3$–$1.4$ TeV assumes decays only to $tZ$, $th$, and $bW$. This paper argues that when a lighter pseudo-scalar $a$ is present, the decay $T\\to t a$ with $a\\to gg$ or $a\\to b\\bar b$ opens a previously unnoticed search gap. Recasting three existing ATLAS and CMS searches for pair-produced top partners, the authors find that for $a\\to gg$ dominance the bound on $m_T$ never passes about $550$ GeV for $m_a<2m_t$, and drops to roughly $390$–$400$ GeV for $m_a\\simeq 50$ GeV. The reason is kinematic: a boosted light $a$ decays into a collimated jet pair that mimics a single QCD jet, so standard top-partner signals are missed.","feed_headline":"LHC bound on top partner can drop to 400 GeV","feed_subtitle":"When the top partner decays through a light pseudo-scalar to gluons, current searches miss it.","key_machinery":"The load-bearing mechanism is the decay chain $T\\to t a$ followed by $a\\to gg$ or $b\\bar b$, with $a$ a light pseudo-Nambu-Goldstone boson coupling to gluons and fermions through the effective Lagrangian (0.1). At low $m_a$ the two gluons from $a$ are boosted into a single jet whenever their angular separation satisfies $\\Delta\\theta<0.2$, so $T\\bar T$ events look like $t\\bar t jj$ and evade vector-like quark searches. The argument is carried by recasting three searches: the CMS excited-top search (grey region), the ATLAS 8 TeV multi-jet RPV SUSY search (red region), and the ATLAS 13 TeV $T\\to th$ broad-band search (blue region), using the signal counting formula $N_s=\\mathcal{L}\\sigma_{T,\\bar p}(m_T)\\sum_{ij}\\epsilon_{ij}(m_T)\\,\\mathrm{BR}_i\\mathrm{BR}_j$.","core_discovery":"The central discovery is that current LHC searches leave composite-Higgs top partners with exotic decays essentially unconstrained in a wide mass window. In the benchmark $T\\to t a\\to t gg$ with $\\mathrm{BR}(T\\to t a)=100\\%$, the pair-production bound falls to $m_T\\gtrsim 550$ GeV for any $m_a<2m_t$, and as low as $m_T\\approx 390$ GeV at $m_a=50$ GeV, compared with the $1300$–$1420$ GeV combined ATLAS bound when only standard decays are assumed. For $T\\to t a\\to t b\\bar b$, the bound is stronger but still drops to about $900$ GeV for light $a$. The paper also shows that when both $th$ and $ta$ decays are present, masses below $800$ GeV remain allowed for $a\\to gg$ as long as $\\mathrm{BR}(T\\to th)\\lesssim 20\\%$.","pith_inferences":["A direct consequence the authors only hint at: naturalness arguments for a top-partner regulator around $1$ TeV are not ruled out by current data if the exotic channel dominates.","The same 'merged dijet' loophole likely weakens bounds on other resonances decaying to gluons when produced in association with tops; a generic $t\\bar t+X$ recast with $m_{jj}$ tagging would test it.","A testable extension: rerun the CMS excited-top analysis with actual jet-substructure information instead of the $\\Delta\\theta<0.2$ single-jet assumption; measuring the tagging efficiency of $a\\to gg$ directly would move the mass bounds by hundreds of GeV."],"forward_implications":["The standard ATLAS/CMS lower bound $m_T>1.3$–$1.4$ TeV applies only within the assumption of standard decays; a composite-Higgs top partner with $T\\to t a\\to t gg$ can be as light as about $400$ GeV.","For $a\\to gg$, no $m_a<2m_t$ is excluded above about $550$ GeV by QCD pair production, leaving a broad band of viable parameter space.","For $a\\to b\\bar b$, the $T\\to th$ search regains sensitivity and exclusions above $1$ TeV reappear for $m_a\\gtrsim 75$ GeV.","Lowering the summed-jet-mass cut in the 13 TeV RPV SUSY search, or adding a dedicated low-mass dijet resonance tag, would close much of the gap.","The inclusive and exclusive rescaling formulae give a simple way to translate standard search bounds into the mixed-decay case when searches have no sensitivity to $ta$."],"supporting_citations":[{"why":"Supplies the $T\\to t a$ interaction and shows that exotic branching ratios can be sizeable in partial-compositeness models.","marker":"[17]"},{"why":"The ATLAS combination that sets the standard 1.30–1.42 TeV bound the paper bypasses.","marker":"[12]"},{"why":"The ATLAS 13 TeV $T\\to th$ broad-band search whose 1-lepton regions are recast to give the blue $a\\to b\\bar b$ exclusion.","marker":"[6]"},{"why":"The CMS excited-top search recast to cover very light $a$ when its decay products merge into a single jet.","marker":"[46]"},{"why":"The ATLAS 8 TeV multi-jet RPV SUSY search recast to produce the red region, strongest for $a\\to b\\bar b$.","marker":"[47]"},{"why":"Establishes that $a\\to gg$ dominates below the $t\\bar t$ threshold in realistic underlying models, making the gap the norm.","marker":"[36]"},{"why":"Shows the low-mass window $15\\lesssim m_a/\\mathrm{GeV}\\lesssim 65$ where direct $a$ searches are weak, motivating $m_a=50$ GeV.","marker":"[38]"},{"why":"The 13 TeV RPV search whose high summed-jet-mass cut rejects the signal, explaining why the weaker 8 TeV search is the relevant recast.","marker":"[48]"}],"fun_headline_variants":["Exotic top partner decays drop mass bound to 400 GeV","Search gap lets top partners hide at 400 GeV","Pseudoscalar decays weaken top partner mass limits","Exotic decay path hides top partners down to 400 GeV"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The low-mass window ($m_T\\sim 400$ GeV at $m_a=50$ GeV) rests on the assumption that an $a\\to gg$ pair with angular separation below 0.2 is always reconstructed as a single jet and that the top-partner signal has the same efficiency as the excited-top search assumes; if real detectors merge less efficiently, or the efficiencies differ, this allowed point can disappear.","fun_headline_variants_meta":{"raw":{"variants":["Exotic top partner decays drop mass bound to 400 GeV","Search gap lets top partners hide at 400 GeV","Pseudoscalar decays weaken top partner mass limits","Exotic decay path hides top partners down to 400 GeV"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000405,"raw_usage":{"total_tokens":2079,"prompt_tokens":887,"completion_tokens":1192,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":503,"completion_tokens_details":{"reasoning_tokens":1125}},"tokens_in":503,"tokens_out":1192,"duration_ms":8456,"temperature":1.0,"reasoning_tokens":1125,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:06:31.610066+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take $m_a=50$ GeV and $\\mathrm{BR}(a\\to gg)=1$, and ask whether the CMS 13 TeV excited-top dataset excludes $T\\bar T$ with $m_T=400$ GeV once the $a\\to gg$ system is simulated with full shower and detector response rather than the $\\Delta\\theta<0.2$ single-jet assumption. A 95% CL bound excluding $m_T=400$ GeV would falsify the paper's headline gap; an observed signal would confirm it.","supporting_citations":[{"cited_title":"Vector-like quarks coupling discrimination at the LHC and future hadron colliders","cited_arxiv_id":"1710.02325","evidence_quote":"The ATLAS 8 TeV multi-jet RPV SUSY search recast to produce the red region, strongest for $a\\to b\\bar b$."}],"review_version":1}