跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2610.40749 · v2

Combinatorial Residual Network: a decay-chain-aware classifier without explicit jet assignment

首次提交 2026-10-30 00:04
版本更新 2026-11-17 00:43
雷达观测 2026-11-17 16:10
arXiv ↗PDF ↗

摘要

Final states with several jets from the decays of heavy particles pose a combinatorial problem: the jets must be assigned, at least implicitly, to their parent particles. The Combinatorial Residual Network (CoReNet) is a classifier that encodes the assumed decay chain in its architecture, building the event representation level by level along the decay hierarchy and combining all jet assignment hypotheses, evaluated with shared weights, with learned weights. The exchange symmetries of the decay are built in, the network size does not depend on the number of hypotheses, and no generator-level jet assignments are used in training. Developed for the bbˉbbˉb\bar{b}b\bar{b} final state and used in several CMS searches, CoReNet with 1508 parameters performs well above kinematic baselines on fast-simulated Higgs boson pair production against QCD multijet events and trains in minutes on a single GPU, and its hypothesis weights select the truth-matched jet pairing in most signal events in which it is defined. Since only the encoded decay chain is specific to the final state, we test the construction on HH→bbˉW+W−HH \to b\bar{b}W^+W^- against semileptonic ttˉt\bar{t}, where encoding the decay chain of the background as well improves the classification at a cost of about 10310^{3} parameters. In both final states, CoReNet performs comparably to a state-of-the-art transformer-based network more than two orders of magnitude larger. The implementation is publicly available at https://github.com/cms-cmu/CoReNet.

作者备注:24 pages, 7 figures, 6 tables. Code available at https://github.com/cms-cmu/CoReNet

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v2,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-17 00:43 · 最新观测版本

Combinatorial Residual Network: a decay-chain-aware classifier without explicit jet assignment

Final states with several jets from the decays of heavy particles pose a combinatorial problem: the jets must be assigned, at least implicitly, to their parent particles. The Combinatorial Residual Network (CoReNet) is a classifier that encodes the assumed decay chain in its architecture, building the event representation level by level along the decay hierarchy and combining all jet assignment hypotheses, evaluated with shared weights, with learned weights. The exchange symmetries of the decay are built in, the network size does not depend on the number of hypotheses, and no generator-level jet assignments are used in training. Developed for the bbˉbbˉb\bar{b}b\bar{b} final state and used in several CMS searches, CoReNet with 1508 parameters performs well above kinematic baselines on fast-simulated Higgs boson pair production against QCD multijet events and trains in minutes on a single GPU, and its hypothesis weights select the truth-matched jet pairing in most signal events in which it is defined. Since only the encoded decay chain is specific to the final state, we test the construction on HH→bbˉW+W−HH \to b\bar{b}W^+W^- against semileptonic ttˉt\bar{t}, where encoding the decay chain of the background as well improves the classification at a cost of about 10310^{3} parameters. In both final states, CoReNet performs comparably to a state-of-the-art transformer-based network more than two orders of magnitude larger. The implementation is publicly available at https://github.com/cms-cmu/CoReNet.

查看此版本解读 →查看此版本原文 ↗