[论文解读] Reconstructing Toponium using Recursive Jigsaw Reconstruction

Aman Desai, Amelia Lovison+1 more|arXiv (Cornell University)|Jan 26, 2026

Image Processing and 3D Reconstruction被引用 0

一句话总结

论文展示了在 t t̄ 阈值处通过 Recursive Jigsaw Reconstruction 重建 toponium，并引入两个变量以在 dileptonic t t̄ 事件中提升信号区分度。

ABSTRACT

The results from the ATLAS and CMS experiment at the Large Hadron Collider indicate the existence of a top-quark pair bound state near the $ tbar$ threshold region. We present a method relying on Recursive Jigsaw Reconstruction to reconstruct the toponium bound state at the $ tbar$ threshold region. We propose incorporating two variables in the analysis that can improve sensitivity to the toponium signal. Our results indicate that this method may be useful to gain additional insights into the physics phenomenology of the $ tbar$ threshold region.

研究动机与目标

• Motivate the search for a top-quark pair bound state (toponium) near the t t̄ threshold observed by ATLAS and CMS.
• Develop a reconstruction strategy for dileptonic t t̄ final states despite two neutrinos in the final state.
• Apply Recursive Jigsaw Reconstruction to distinguish toponium signal from t t̄ background.
• Introduce new kinematic variables to improve signal sensitivity in the threshold region.

提出的方法

• Use Recursive Jigsaw Reconstruction within the RestFrames framework to build a decay tree for t t̄ → bbW(lν)W(lν) with visible and invisible final-state particles.
• Compare four constraining reconstruction schemes (A: Mtop^a = Mtop^b, B: MW^a = MW^b, C: min Σ Mtop^2, D: min ΔMtop) and select the optimal one (A).
• Define two novel observables, Δφ(t t̄) and Nc_hel, to distinguish signal from background in dileptonic t t̄ events.
• Construct a 3x3-region phase space grid in Δφ(t t̄) and Nc_hel and evaluate significance S/√(S+B) using Mt t̄ as the mass observable.
• Assess performance with Monte Carlo samples tuned to LHC Run 3 (√s = 13.6 TeV, 300 fb⁻¹) and toponium signal modeled via NRQCD-based reweighting.

Figure 1 : Example decay tree diagram of the process $t\bar{t}\rightarrow b\bar{b}W(l\nu_{l})W(l\bar{\nu_{l}})$

实验结果

研究问题

• RQ1 Can Recursive Jigsaw Reconstruction improve the reconstruction of dileptonic t t̄ final states with two neutrinos?
• RQ2 Do new kinematic variables Δφ(t t̄) and Nc_hel provide enhanced discrimination between toponium and t t̄ backgrounds at the t t̄ threshold?
• RQ3 What is the achievable significance for toponium signal in the Run 3 configuration using the proposed method?
• RQ4 Which reconstruction constraint yields the most consistent top mass in the dilepton t t̄ system?

主要发现

• The A reconstruction constraint (Mtop^a = Mtop^b) yields the most consistent top mass distribution among the four methods.
• The variables Δφ(t t̄) and Nc_hel show distinctive signal vs. background behavior at parton level, aiding discrimination.
• Nine phase-space regions are explored to optimize signal significance.
• The optimal region Δφ(t t̄) ∈ [-2, 2] and Nc_hel ∈ [0.4, 1] yields a reconstructed t t̄ mass distribution with a significance of 15.3σ.

Figure 2 : Invariant mass distributions of the top-quark pair as evaluated by the reconstruction algorithms.