Freya Blekman

different signatures tau leptons can leave in CMS

Tau leptons are electrons but heavier, but unlike electrons they are unstable and create many different detector signatures. So, #machinelearning This #CMSpaper describes how we use convolutional #neuralnets (that's ones that do image recognition!) to better see tau leptons arxiv.org/abs/2511.05468

definitions of sub-jets and a way to quantify jets called the "Lund plane"

How does a bottom #quark become a spray of many composite particles called a bottom jet? This is actually a very difficult and open research question, and this #CMSPaper tries to measure that jet fragmentation process in heavy ion collisions arxiv.org/abs/2511.10666

Join me on a tour of the CMS Experiment at CERN with Prof. Freya Blekman. We head 100m underground to visit this particle physics experiment on the Large Hadron Collider, take a closer look at the… Visiting the competition at #CERN

Want to know what a CMS underground visit is like? Earlier this year, I took particle physicist (and science communicator extraordinaire) Clara Nellist to visit the Compact Muon Solenoid underground at CERN. The result is now on YouTube!

youtu.be/dumEjuLWo3Q?...

diagram explaining how the Higgs boson would decay to Dark QCD signatures

This #CMSPaper looks if there are Higgs bosons decaying to very light "dark QCD" particles hiding in Higgs decays, which would show up as jets of particles that are inconsistent with the QCD of the standard model arxiv.org/abs/2511.11888

The CMS pixel detector and its support structure show up beautifully when studying the backgrounds for this paper

Are there undiscovered particles being produced at the LHC that we just missed? This #CMSPaper looks for such long-lived particles. Understanding the detector is very important there as known particles can make that signature when they traverse the material of our detector arxiv.org/abs/2511.08212

dimuon invariant mass distribution. No significant bumps to be seen.

Are there extra Z-boson like undiscovered particles made at higher mass? Or did we miss them because they can only see them when they're made together with quarks? This #CMSPaper scans the dimuon spectrum - it is a #nullresult (no extra Z' bosons spotted!) arxiv.org/abs/2511.11853

I will convey the message:-)

Ah, Sofia did that project with a colleague so that I also can’t read. Let’s just say that I’m a stricter boss and from my group plots with unreadable labels tend to not go out). The property on the y axis is the angle between two objects in the detector in a collision. But I can’t read which either

I don’t know what the plot is that you talk about but often the y axis is then the number of collisions that have that property (so the plot is a histogram) or a different property of the same collision (then the plot is measuring the correlation between those properties), typically a scatter plot

E_T^{miss}
: Sofia is using latex notation but missing transverse energy (= missing E-T = E_T^{miss}) is a number that quantifies how much energy is unaccounted for in a single LHC collision.

Deze insta thread is mss nog nuttiger www.instagram.com/p/DU8E80SCHf...

Here is an intro to LHC analysis bsky.app/profile/frey...

In quantum physics, the decay width is a fundamental parameter that quantifies a particle's instability. It is also a way to check if a particle is also decaying to undiscovered particles when the decay width is different from what can be calculated using known particles.

Machine learning output isolating off-shell Higgs bosons in the WW signature (its super rare)

This #CMSPaper measures the decay width of the Higgs boson in a single signature, which involves a trick comparing Higgs bosons in their on-shell and off-shell production, which means this paper isolates these separately using machine learning arxiv.org/abs/2601.05168

But my point is really that most of these graphs are completely different so the explanation would also be completely different

maybe a video on what is involved with doing LHC physics as far as statistics and interpretation is not a bad idea for Instagram or TikTok. Over the years, I have done a few 20+ post threads here discussing the intricacies of important results. But that is not really doable for every single paper

It really depends on the figure. Some simply count collisions. Some measure properties of many collisions, sometimes after many many years of machine learning/AI and further analysis. Some compare different physics interpretations after years of data and analysis. Some show the performance of tools

Selfie of Freya. On a train.

Greetings from the train to my vacation 🎿

Still finishing some urgent things that affect people, like grant reviews and reference letters, before going offline fully. Don’t worry, through the powers of internet scheduling I might still be posting about every #cmspaper

Each of those are a figure from the science paper I discuss in those posts. Usually the alt text gives more info, but as every paper is different, so is every figure 🤓

But I’m always happy to answer questions

Want to know more about what particle physics is like, told by others other than only me? Our group's Instagram account gives a voice to a very wide group of scientists

For example, in this video, Lucia shows what it's like to take data with the CMS experiment www.instagram.com/reel/DUxjg9-...

there are some excesses in this analysis by the way, but nothing too exciting yet. Stay tuned, we have a lot more data in hand!

Are there undiscovered charged heavy Higgs bosons? If they are heavier than the top quark, they would decay into a top quark + b quark signature. This #CMSPaper looks for these, and also compares them in dedicated theory frameworks that include neutral heavy Higgs bosons arxiv.org/abs/2512.24471

Diphoton invariant mass

This #CMSPaper searches for a heavy particle decaying into pairs of undiscovered light bosons, which in turn decay into photons that can partially overlap. This is technically a difficult signature because overlapping photons are not easy to disentangle arxiv.org/abs/2601.00183

Comparison of different (supervised) machine learning searches

One of the classical ways to look for undiscovered particles at the LHC is to look for unexpected resonances in the jets coming from quarks and gluons. This #CMSPaper compares the cutting edge of #machinelearnining #ai methods to see how well they do for top quark resonances arxiv.org/abs/2512.20395

distribution of Z bosons and photons.

This #CMSPaper measures the simultaneous production of Z bosons and photons. That way their interactions can be measured and can be compared to predictions by the standard model arxiv.org/abs/2512.08582

Comparison of the photons to the standard model, in the Zeppenfeld variable

This #CMSPaper looks for production of photons in these boson collisions, meaning the LHC is a W boson collider, which is the first time this has ever been seen! It measures properties of the photons produced and compares them to the standard model predictions arxiv.org/abs/2512.00502

scanning the four-lepton invariant mass spectrum, no bumps visible

The Higgs boson can decay to many different particles from the standard model. But... is the Higgs boson also decaying to undiscovered particles? This #CMSPaper looks for extra (light) bosons that would create the Higgs boson to four electron signature arxiv.org/abs/2511.19563

Deep into the detector 👷

Today’s #PhotoOfTheWeek features the @atlasexperiment.bsky.social detector during the year-end technical stop (YETS) – a brief winter switch-off that allows for essential maintenance and upgrades on equipment and detectors, preparing them for the following physics run.

Lustig, wie doll meine Ausstiegsverkündung einzelne Profs triggert, mir zu sagen, ich sei ja ganz offensichtlich ungeeignet für Wissenschaft. Als ob man mit jemandem Schluss macht & derjenige gar nicht schnell genug beteuern kann, die Beziehung ja selbst nicht mehr gewollt zu haben. 😂 #IchBinHanna

starting out by citing Kaku I would not take this very seriously.

Gravitons exist in all sorts of grand unification physics theories. This #CMSPaper looks for them in the photon+missing energy signature, which is also a good signature to check for #darkmatter by the way. It looks in the 2016 through 2018 CMS data, and is a #nullresult arxiv.org/abs/2511.17310