Searches in CMS for New Physics in Final States with Leptons

This paper presents recent CMS results from Run-II LHC data searching for heavy mediators predicted by various new physics models in final states containing leptons.

The Gist

Imagine the Large Hadron Collider (LHC) as the world's most powerful particle smasher. It takes tiny building blocks of the universe (protons) and smashes them together at nearly the speed of light to see what happens. Usually, these collisions create a chaotic mess of particles, but physicists are looking for something specific: leptons.

Think of leptons (electrons, muons, and taus) as the "VIPs" of the particle world. Unlike other particles that get lost in the noise of the collision, leptons are clean, easy to spot, and leave a clear trail. If there is "New Physics" hiding behind the Standard Model (our current rulebook for how the universe works), it's likely to show up in the VIP section.

This paper is a report from the CMS experiment (one of the giant detectors at the LHC), summarizing five different "treasure hunts" they conducted to find these VIPs. They looked for clues that might reveal hidden universes, dark matter, or new forces.

Here are the five hunts, explained with simple analogies:

1. The "Ghostly" Hunt (Soft Leptons & Supersymmetry)

• The Theory: Some theories suggest that for every heavy particle we know, there is a lighter, "cousin" particle (like a chargino or neutralino) that is almost the same weight.
• The Challenge: When these cousins decay, they don't throw a big party; they whisper. They produce "soft" leptons—particles with very low energy that are usually too weak to be seen by standard detectors. It's like trying to hear a whisper in a hurricane.
• The Solution: The team built a special "super-hearing aid" (dedicated reconstruction) to catch these faint whispers, even if they are moving very slowly or appear slightly later than expected (displaced).
• The Result: They didn't hear the whisper. However, by listening this closely, they managed to rule out a whole range of "ghost" particles that previous experiments missed, effectively closing a gap in our knowledge.

2. The "Hidden Door" Hunt (Higgs-Portal Light Scalars)

• The Theory: The Higgs boson (the particle that gives mass to others) might be a "portal." Instead of just decaying normally, it might secretly open a door to a hidden world of "light scalars" (tiny, invisible particles) that then turn into pairs of muons and other particles.
• The Challenge: The background noise (QCD multijets) is like a crowded room where everyone is shouting. Finding a specific conversation is hard.
• The Solution: They used a clever trick: looking for a "diagonal match." They checked if the mass of the muon pair matched the mass of the other particles perfectly. This acted like a noise-canceling headphone, filtering out 96% of the background noise.
• The Result: No hidden doors were found. But they set strict rules on how often this "secret door" could be opening, pushing the limits of what we know about the Higgs boson.

3. The "Low-Altitude" Hunt (Scouting for Light Resonances)

• The Theory: There might be new, light particles (resonances) that decay into pairs of tau particles (heavy cousins of electrons).
• The Challenge: Standard detectors have a "speed limit" (trigger threshold). They ignore anything too light or slow to save storage space, like a bouncer at a club who only lets in VIPs with expensive tickets. This means low-mass particles (20–60 GeV) were being ignored.
• The Solution: They used "Scouting Data." Imagine a bouncer who, instead of checking IDs, just takes a quick photo of everyone passing by, even the ones without tickets. This allowed them to look at the "low-mass" crowd they usually ignore.
• The Result: They scanned the low-mass crowd and found nothing unusual. But this was the first time the LHC successfully looked in this specific low-mass zone, proving the "scouting" method works.

4. The "One-Way Ticket" Hunt (Scalar Leptoquarks)

• The Theory: Leptoquarks are hypothetical particles that act as bridges, connecting quarks (which make up protons) and leptons. Usually, we look for them appearing in pairs. But some theories say they can appear alone, created when a muon smashes into a quark.
• The Challenge: This is a rare, single-production event, like finding a specific needle in a haystack where the needle is made of the same material as the hay.
• The Solution: They used a "Smart Filter" (a machine learning algorithm called a Boosted Decision Tree) to sort through millions of collisions, looking for a high-energy muon and a jet of particles that fit a specific pattern.
• The Result: No leptoquarks were found. However, they proved that if these particles exist, they must be incredibly heavy (up to 5 TeV), effectively ruling out the lighter versions.

5. The "Ultra-Light" Hunt (Axion-Like Particles)

• The Theory: There might be "Axion-Like Particles" (ALPs)—extremely light, ghostly particles that the Higgs boson could decay into. These would then turn into four electrons.
• The Challenge: These particles are so light and decay so quickly that the electrons they produce look like a single blob or a confused mess in the detector. It's like trying to identify two people holding hands while they are running so fast they look like a blur.
• The Solution: They developed a "Merged Object" technique. Instead of trying to separate the blur, they trained a computer to recognize the specific "blur signature" of two electrons coming from the same tiny point, distinguishing them from random background noise.
• The Result: No ALPs were found. But they set the first-ever limits on these particles in this specific mass range, showing that the Higgs boson isn't hiding them there.

The Big Picture

In all five of these hunts, no new physics was found. The universe, so far, is behaving exactly as the Standard Model predicts.

But here is the good news:
Think of these results not as "failures," but as tightening the net. By proving that these specific types of new particles don't exist in these specific ranges, the scientists are narrowing down where the real "New Physics" might be hiding. They have also proven that their new tools (better hearing aids, scouting data, and smart filters) work perfectly.

As the LHC prepares for its next, even more powerful run (the High-Luminosity LHC), these experiments have laid the foundation. They are ready to look even deeper, listen even quieter, and find whatever is hiding in the shadows of the lepton world.

Technical Summary

Based on the provided paper, here is a detailed technical summary of the CMS searches for New Physics in final states with leptons.

1. Problem Statement

The Standard Model (SM) of particle physics, while successful, fails to explain several fundamental phenomena, including dark matter, matter–antimatter asymmetry, and the hierarchy of mass scales. Many theoretical extensions (e.g., Sequential Standard Model, Grand Unified Theories, Extra Dimensions, Leptoquarks, Vector-like Leptons) predict heavy mediators at the TeV scale or light, weakly coupled particles.

Leptons (electrons, muons, taus) offer "clean" experimental signatures with precise momentum resolution and reduced QCD backgrounds, making them ideal probes for these new physics (NP) scenarios. However, specific NP models present unique challenges:

• Compressed Spectra: Small mass differences between new particles and the Lightest Supersymmetric Particle (LSP) result in very soft (low momentum) leptons that are difficult to trigger on.
• Displaced Vertices: Long-lived particles decay away from the primary interaction point, requiring specialized reconstruction.
• Low-Mass Resonances: Standard triggers often discard low-mass events ($< 20$ GeV), creating a blind spot for light scalars.
• Exotic Decays: Rare decays like $H \to aa \to 4e$ involve ultra-light particles and complex reconstruction challenges (e.g., merged electron pairs).

2. Methodology

The analysis utilizes proton–proton collision data from the CMS experiment at the LHC.

• Datasets: Primarily Run-II data ($\sqrt{s} = 13$ TeV, $138 \text{ fb}^{-1}$) and, for specific low-mass searches, early Run-III data ($\sqrt{s} = 13.6$TeV,$61.9 \text{ fb}^{-1}$).
• Analysis Strategies: Five distinct analyses cover different mass regimes and production mechanisms:
  1. Soft Leptons (EXO-23-017): Targets nearly mass-degenerate charginos/neutralinos. Uses dedicated reconstruction to lower $p_T$ thresholds ($e \approx 1$ GeV, $\mu \approx 3.5$ GeV) and searches for both prompt and displaced signatures (up to 10 cm). Backgrounds (Drell–Yan, top, non-prompt) are estimated via data-driven template fits.
  2. Higgs-Portal Light Scalars (EXO-24-034): Searches for $H \to SS \to (\mu^+\mu^-)(h^+h^-)$. Events are reconstructed with a dimuon pair and two hadronic objects. A novel 2D diagonal mass requirement ($m_{\mu\mu} \approx m_{hh}$) suppresses QCD multijet background by ~96%. Backgrounds are estimated from sidebands using parametric fits.
  3. Inclusive Low-Mass $\tau\tau$ (EXO-24-012): Utilizes the CMS Scouting dataset (high-rate, reduced event info) to access the 20–60 GeV mass range inaccessible to standard triggers. Uses the Hadron-Plus-Strips algorithm with $\pi^0$ recovery and a Deep Neural Network (TauNet) for $\tau$ identification.
  4. Scalar Leptoquarks via $\mu$-quark Scattering (EXO-24-005): Investigates single-production leptoquarks ($\mu q \to LQ$). Events feature a high-$p_T$ muon and a jet. Signal discrimination uses Boosted Decision Trees (BDT) across eight categories based on muon multiplicity, $b$-tag content, and BDT output.
  5. Ultra-light ALPs (EXO-24-031): Searches for $H \to aa \to (ee)(ee)$ with ALP masses of 10–100 MeV. Employs a merged-electron-pair reconstruction technique where two Gaussian-sum-filter tracks are matched to a common electromagnetic supercluster. A BDT distinguishes signal from photon conversions and misreconstructed electrons.

3. Key Contributions

• Extended Kinematic Reach: Lowered lepton $p_T$ thresholds to $\sim 1$ GeV, allowing the exploration of compressed supersymmetric spectra previously unobservable.
• Novel Triggering and Reconstruction:
  • First inclusive LHC measurement of low-mass $\tau\tau$ resonances using scouting data.
  • First CMS limits on sub-GeV scalars from Higgs decays within the tracker volume.
  • Development of merged-electron reconstruction for ultra-light ALPs in Higgs decays.
• New Production Mechanisms: Established sensitivity to single-production leptoquarks via muon-quark scattering, surpassing previous pair-production constraints.
• Displaced Vertex Sensitivity: Comprehensive searches for long-lived particles with lifetimes ranging from 100 $\mu$m to 10 cm across multiple channels.

4. Results

No statistically significant deviations from the Standard Model background expectations were observed in any of the five analyses. Consequently, 95% Confidence Level (CL) upper limits were set:

• Compressed SUSY (EXO-23-017): Excluded higgsino masses up to 140 GeV for mass splittings ($\Delta m$) between 0.6 and 50 GeV, effectively closing the mass gap left by LEP.
• Higgs-Portal Scalars (EXO-24-034): Set limits on the branching fraction $B(H \to SS)$ down to **$10^{-5}$** for scalar lifetimes ($c\tau$) around 1 mm.
• Low-Mass $\tau\tau$ (EXO-24-012): Established upper limits on the cross-section $\sigma(pp \to \phi \to \tau\tau)$ in the 20–60 GeV range, reaching the order of 10 pb.
• Scalar Leptoquarks (EXO-24-005): Excluded scalar leptoquark masses up to 5 TeV for large couplings ($\lambda \approx 1$), extending sensitivity to single-production regimes.
• Axion-Like Particles (EXO-24-031): Set limits on $B(H \to aa \to 4e)$ at the level of $10^{-5}$to$10^{-6}$ for ALP masses between 10 and 100 MeV.

5. Significance

This work demonstrates the CMS Collaboration's ability to probe diverse New Physics scenarios through advanced reconstruction techniques and novel data streams.

• Completeness: The results cover a vast mass spectrum, from MeV-scale (ALPs) to multi-TeV (Leptoquarks), addressing gaps left by previous experiments (e.g., LEP).
• Methodological Innovation: The successful application of scouting data for low-mass resonances and merged-track reconstruction for ultra-light particles sets a precedent for future rare-event searches.
• Future Outlook: These results establish a robust foundation for the High-Luminosity LHC (HL-LHC) era. The upcoming tenfold increase in luminosity, combined with improved tracking and real-time reconstruction, will further enhance sensitivity to soft, displaced, and rare leptonic signatures, potentially uncovering the nature of dark matter or the origin of mass hierarchies.