Geneva / Cambridge, MA — Summer 2026
The Particle Physics Collaboration celebrates a golden era of CMS physics, as the world’s greatest collider powers down for Long Shutdown 3
On 29 June 2026, at 6:00 in the morning, CERN switched off the Large Hadron Collider and stepped into Long Shutdown 3. After circulating its first beams in September 2008, delivering its first collisions in 2009, and powering three remarkably successful runs, the world’s most powerful particle accelerator fell silent—not in defeat, but in preparation for its most ambitious transformation yet. As CERN’s Director-General put it, the coming upgrade is “by far the largest thing CERN has done in the last 20 years.”
For the Particle Physics Collaboration (PPC) at MIT, this is more than a milestone in an accelerator’s logbook. It is the end of a chapter we helped write, one collision, one calibration, and one very stubborn boson at a time.
“The LHC has exceeded every expectation. For nearly two decades, it has transformed our understanding of the Universe and inspired generations of scientists, engineers and citizens around the world.”
— Oliver Brüning, CERN Director for Accelerators and Technology
So, before the champagne goes flat, let’s raise a glass to the best of what happened to the PPC and to CMS during this extraordinary time.
1. The discovery of the century: the Higgs boson
Everything else on this list traces back to a single day. On 4 July 2012, the ATLAS and CMS Collaborations announced the discovery of the Higgs boson, confirming—nearly half a century after it had been proposed—the mechanism responsible for giving elementary particles their mass. It remains the LHC’s defining scientific achievement and the discovery for which the accelerator will be remembered as long as we seek to understand the fundamental laws of nature.
CMS did not discover the Higgs boson by chance. It required a detector built with extraordinary precision, a trigger system capable of selecting a handful of promising events from billions of proton-proton collisions every second, and a worldwide computing infrastructure to process and analyze the resulting torrent of data—the very kinds of systems the PPC group has spent its career designing, building, and operating. The Higgs boson was not simply found; it was painstakingly reconstructed, decay channel by decay channel, from the raw output of one of the most complex scientific instruments ever built.
“On 4 July 2012, we didn’t just add a particle to the table. We closed the last gap in the standard model … and CMS was holding one half of the pen.”
With faculty Christoph Paus at the helm of CMS’s Higgs Physics Analysis group from 2011 to 2012 the PPC directed the analysed that led to the discovery.
- Higgs to two photons turned out to be CMSs’ most significant Higgs boson decay channel for the discovery and graduate student Joshua Bendavid (MIT PhD 2013) designed and implemented the highly optimized and complex analysis based on multivariate techniques, while graduate student Ming Ming Yang (MIT Phd 2015) refined the analysis and published the full Run 1 data set. Postdoc Fabian Stoeckli kept the students at bay and made sure the background model was studied carefully enough to get the sensitivity just right.
- Higgs to 4 leptons known as the golden channel gave the CMS discovery a good push and had various PPC contributions: graduate students Duncan Ralph (MIT PhD 2014) worked on the overall selection optimization, Kevin Sung (MIT PhD 2013) provided the first multivariate muon selection while Si Xie (MIT PhD 2012) the corresponding multivariate electron selection. Postdoc Kristian Hahn did the overseeing.
- Higgs to WW was the final significant channel that would push the CMS discovery over the required ‘significance edge’. Researcher Guillelmo Gómez-Ceballos from MIT was the hard and soul of this analysis. It was the first Higgs analysis the PPC embarked on and was the first to present multivariate techniques (Boosted Decision Trees) to CMS and to the LHC as a whole. Postdoc Marco Zanetti worked with graduate students Kevin Sung and Si Xie (MIT PhD 2012) who provided the leg work for the muon and electron decay channels, respectively.
- Higgs to ditau was the final channel MIT pushed hard on, but due to the limited data sample and the complexity of the analysis the sensitivity to a standard model Higgs boson was not too high. Then junior faculty Markus Klute and his team of postdocs Erik Butz, and Roger Wolf and graduate students Valentina Dutta (MIT Phd 2014) and Duncan Ralph got the analysis completed. This work put CMS in a particularly strong position for the Higgs boson discovery, as ATLAS was unable to complete a comparable analysis in time. The subsequent development of the technique ultimately led to one of CMS’s most elegant results: the observation of the Higgs boson coupling to leptons.
Every W-boson-mass measurement, every dark-matter search, every precision result that followed is, in a sense, a conversation with the particle we first glimpsed that day.
2. The hunt for the dark side: Mono-X on Run-2 data
If the Higgs was the answer to a fifty-year-old question, the Mono-X program was the group leaning hard on the door to the next one: what is dark matter?
Across the Run-2 dataset — the enormous 13 TeV harvest of 2015–2018 — the PPC group carried out a whole series of Mono-X measurements: the search for events in which a single, visible object recoils against nothing at all. A lone jet (mono-jet), a lone photon (mono-photon), a lone Z or W, even a lone Higgs (mono-Higgs) — recoiling against large missing transverse momentum, the tell-tale signature of particles that sail straight through the detector without leaving a trace. If dark matter is produced at the LHC, this is precisely how it announces itself: by its absence.
These analyses are among the most demanding in experimental particle physics. Their signature is missing energy, meaning the challenge is to understand—with extraordinary precision—every ordinary process that can mimic it: neutrinos, mismeasured jets, and instrumental effects. The Mono-X program became a showcase of the PPC group’s strengths: rigorous background modeling, sophisticated data-driven techniques, and meticulous control of systematic uncertainties. No evidence for dark matter has emerged so far, but each new measurement has narrowed the range of viable possibilities, setting some of the world’s most stringent constraints and steadily guiding the search toward whatever form dark matter ultimately takes.
“You point the whole detector at a fistful of nothing and ask, very carefully, whether that nothing is heavier than it should be. That is the Mono-X game — and nobody plays it more carefully than this group.”
- Mono-Z(ll) – Graduate student Dylan Hsu (PhD 2019) got the foot in the door with a leptonic Z boson recoiling against nothing. The analysis was an outgrowth and tight collaboration of Research Scientist Guillelmo Gomez-Ceballos detailed cross section analysis of single and double vector boson production.
- Mono-Photon – Postdoc Yutaro IIyama got the analysis going and was eventually joined by graduate student Brandon Allen (MIT PhD 2019), who completed his thesis on this topic.
- Mono-Jet – Postdoc Zeynep Demiragli made the monojet search her life goal and got this complex analysis including jet substructure with hadronic W and Z decays published on the initial Run-2 data while eventually our prolific postdoc Zhangqier Wang completed the legacy Run-2 Mono-Jet analysis. The publication also set a temporary end to the PPC’s Mono-X effort for the PPC. We might come back to it for the full Run 3 data.
- Mono-Top – Graduate student Siddharth Narayanan (MIT PhD 2019) developed the first deep learning neural net tagger together with postdoc Benedikt Maier using sophisticated jet substructure variables like energy correlations to search for Mono-top events.
- Mono-Higgs – Postdoc Benedikt Maier carried the deep learning idea further to look for Higgs decaying to two b quarks possibly in a single fat jet with substructure.
But nature refused to be that tidy, because none of the analysis could pin down even a hint of dark matter. But, if dark matter lives in a whole dark sector, its own family of particles and forces, only faintly touching ours, then it need not leave the detector as pure missing energy at all. So the group cast the net wider. The same instincts that powered Mono-X grew into a program of dark-sector searches: hunts for dark photons, feebly-coupled cousins of the ordinary photon that would surface as narrow resonances or displaced signatures, and, most strikingly, searches for dark showers like Soft Unclustered Energy Patterns (SUEP). In a SUEP event, a strongly-coupled hidden sector fragments into dozens or hundreds of soft particles sprayed nearly isotropically across the detector — the polar opposite of a clean, collimated jet. It is arguably the least jet-like signature at the LHC, and finding it demands throwing out the standard reconstruction playbook and inventing new observables from scratch.
“Dark matter didn’t sign a contract to show up only as missing energy. If it hides in a dark sector, we have to be ready for anything … a stray dark photon, or a soft, spherical spray of particles that looks like nothing we searched before. So we widened the net.”
Together, the Higgs and this dark-sector program bracket the two great questions of the LHC era: what gives matter its mass? … DONE! … and what is the dark matter that outweighs it five to one? … still open, but cornered a little more with every result, from just a single photon search in the detector to the messiest dark shower.
3. Rare decays: precision as a discovery tool
Not every search for new physics needs a spectacular signature. Sometimes the most powerful probe is a decay so rare that the Standard Model predicts it will happen only a handful of times in a billion — and then counting, very carefully, whether nature agrees.
The group threw itself into exactly this kind of rare flavor physics: processes in which quarks change flavor through quantum loops so strongly suppressed in the Standard Model that any new heavy particle—even one far too massive to be produced directly at the LHC—could reveal itself through a subtle modification of the decay rate. A textbook example is the decay of a B meson into a pair of muons, an extraordinarily rare process hidden within the overwhelming background of ordinary proton-proton collisions. Isolating it cleanly requires exceptional trigger performance, precise muon reconstruction, and meticulous background modeling. Once measured, its branching fraction can be compared with the Standard Model prediction, turning the decay into an indirect probe of energy scales far beyond the collider’s direct reach.
Turning precision into a discovery tool is the beauty of the flavor frontier. A branching fraction measured to a few percent, sitting exactly where the Standard Model says it should — or, tantalizingly, a hair off — speaks about physics at scales far beyond the beam energy. Every rare-decay result the group produced was another door quietly closed on whole classes of new theories, or a hint kept honest.
“You don’t always need a bang. Sometimes the loudest thing in physics is a decay that happens exactly as often as predicted — three times in a billion, not two, not four — and every model that said otherwise dies on the spot.”
4. Run 3 crossed the finish line — in style
On 19 May 2026, the proton–proton collisions of Run 3, begun way back on 5 July 2022, officially came to an end. And the machine did not coast to the finish. In its final year, the LHC beat its planned data delivery by nearly 10 percent, all while holding the low-instantaneous-luminosity, high-quality running that CMS had specifically requested. Since Run 1 began in 2010, the LHC has handed CMS more than half an inverse attobarn of data — a genuinely staggering pile of physics.
The timing was not an accident, either. By late 2024, both CMS and ATLAS were seeing radiation damage creep into their inner detectors. Knowing exactly when to call it — squeezing out every last good collision without cooking the silicon — was a decision that mattered. Run 3 landed that plane beautifully.
“Run 3 was not just a success — it was a masterclass in knowing your detector, knowing your machine, and asking for exactly the beam you needed. CMS got the data it wanted, in the shape it wanted it.”
— the mood in the PPC offices, May 2026
5. Nature spoke — and it said “W”
Then came the crown jewel. In April 2026, Nature published the CMS measurement of the W boson mass — complete with an admiring commentary. To appreciate how rare that is: this was only the fourth CMS paper in Nature out of more than a thousand CMS publications.
And the result? It is the most precise W-mass measurement from any LHC experiment, and it agrees perfectly with the Standard Model prediction — laying to rest the tantalizing tension raised by the earlier CDF result.
This was a PPC production through and through:
- Joshua Bendavid (MIT PhD 2013) drove the momentum-scale calibration and the software optimization that made the whole analysis tractable.
- Kenneth Long wrangled the theoretical uncertainties hand-in-hand with the theorists (now carrying the torch onward at CNRS, Lyon).
- Jan Eysermans calibrated the hadronic recoil and was the champion of the 2026 low-luminosity running that keeps this program alive.
- Tianyu (Justin) Yang (MIT Phd 2024) owned the PDF studies — and defended his thesis before the first public announcement in September 2024.
“It is a rare occurrence that CMS publishes in Nature. To do it with the most precise W mass in the world, and to have it agree with the Standard Model to the letter — that is a monumental effort finding its reward.”
— from the PPC announcement, “It’s official – Nature has spoken”
First presented to the world in September 2024, then enshrined in the pages of Nature in 2026. Not bad for a boson.
6. A typical CMS Week 2026: the group showed up
The activity of the PPC can be ilustrated by a fairly typical CMS Week in June 2026, with the collaboration taking stock of Run 3 and staring down the Phase-2 upgrades, the PPC delegation was everywhere:
- Zhangqier Wang presented the comparison of dimuon and ditrack reconstruction efficiencies across all of Run 3 — “an important input for a number of ongoing analyses.”
- Jan Eysermans laid out the status of the low-pileup physics program using the fresh 2026 dataset, pointing straight at the next generation of precision measurements — including new ways to pin down the W boson mass.
- The group pushed forward on diboson polarization in Run 3, edging toward the holy grail of longitudinal vector-boson scattering.
And there was hardware, too: Kirill Ivanov picked up a CMS Award for his contributions to the PPD program. Meanwhile the collaboration quietly rolled past its 1500th publication — a number that would have seemed absurd when the first beams circulated in 2008.
“Precision is a team sport. You measure efficiencies, you calibrate recoil, you chase down the last picobarn of piled-up data — and one day it all shows up as a line in Nature.”
7. The next generation is already outrunning us
If you want to know whether a group has a future, look at its students.
In May 2026, Charlotte Myers — one of our own recent UROP students — was named a 2026 Barry Goldwater Scholar, one of only three from MIT. She had thrown herself into a Future Circular Collider workshop in January 2024, spent the following summer doing CMS Higgs rare decay analysis at CERN with the rest of our crew at CERN (and, yes, found time to go sailing), and left a lasting impression on everyone here.
“We knew she was going to go far.”
— Christoph Paus, PPC
8. A bittersweet farewell in the family
The era did not close without loss. Within a few short years, both fathers of the mechanism that gave the LHC its greatest prize left us.
In April 2024, Peter Higgs — the man whose name the boson carries — died at 94. He had waited nearly fifty years, from a 1964 paper to a Geneva auditorium in 2012, to see his prediction confirmed; the shy, modest physicist famously slipped away from the spotlight even as the world celebrated his particle. That CMS was one of the two experiments to hand him that confirmation in his lifetime is a source of quiet pride here.
Then, in June 2026, François Englert — Nobel laureate, honored “for the theoretical discovery of a mechanism that contributes to our understanding of the origin of mass” — died at 93. Together with Higgs (and the late Robert Brout), he gave us the mechanism that lends the W and Z their mass. The particle CMS spent so much of the LHC era measuring carries their fingerprints, and measuring the W mass to unprecedented precision is, in its own way, a tribute to the theory they built.
“They wrote the equations in the 1960s and waited half a century for a machine worthy of testing them. We were lucky enough to be the ones running it.”
What comes next
The LHC is off, but CERN is anything but idle. Over the next years, 1.2 km of magnets and components will be pulled out of the LHC alone, and ATLAS and CMS will be rebuilt into effectively new detectors — all-silicon trackers with billions of channels, timing detectors resolving to tens of picoseconds, and calorimeters that fire at megahertz rates. They’ll have to, because the High-Luminosity LHC (first beams targeted for 2030) will pile up 140–200 collisions per bunch crossing and force the experiments to pick the best events out of five billion interactions every second.
“The LS3 represents a huge and complex logistical and engineering undertaking. In the LHC alone, 1.2 km of magnets and components will be removed and replaced with new equipment.”
— Jean-Philippe Tock, Head of the LS3 Coordination Team
And in May 2026, Europe set its course for the decades beyond: after two years of deliberation and 260+ submissions, the CERN Council chose the electron–positron Future Circular Collider (FCC-ee) — a 90-kilometer ring built to study the Higgs in exquisite detail — as its next flagship, with a funding decision expected by 2028. The very workshop that sent Charlotte Myers to CERN was about that machine. The future is already recruiting.
The toast
So here’s to Run 3, brought to a spectacular conclusion. To the most precise W boson mass ever measured. To a Nature paper, a CMS Award, a Goldwater Scholar, and more than 1,500 publications. To the students who will build the FCC. And to the LHC—the discovery machine that gave us the Higgs boson on 4 July 2012 and has never stopped expanding our understanding of the Universe.
The beam is off. The physics is not.
“Today we say goodbye to the LHC as we have known it, while preparing to welcome its successor. This scientific adventure will extend far into the future.”
— Oliver Brüning, CERN
See you in Run 4.
— The PPC Group, MIT