Chris G

Bikes and sunshine at Munich Institute for Astro- Particle- and Bio Physics

Breaking: It has stopped raining in Munich.

🚀 New Rivet 4.1.1 release!

🔹 Remote HepMC source support (e.g. xrootd)
🔹 Differential negative-weight plots via MC_WEIGHTS
🔹 Runtime toggle for event smearing
🔹 ONNX inference supports both float & double
🔹 Improved DileptonFinder + DIS projection robustness
🔹 New & updated LHC/HERA analyses!

Thanks to all who joined! 👏
If you’re a PhD student or early-career postdoc working with Monte Carlo event generators, keep an eye out for next year’s school—you won’t want to miss it.

#MCnet #HEP #ParticlePhysics #MonteCarlo

A group photo of the participants of the 2025 MCnet school.

📸 A fantastic week at the MCnet Summer School 2025 in Stirling!

From matrix elements and parton showers to neutrino generators, ML, HPC and Public Health—great lectures, hands-on tutorials, and plenty of good conversations (and Scottish weather).

🚀 CONTUR 3 is here!

CONTUR uses RIVET and its library of LHC measurements to test new physics models via fast limit-setting and consistency checks.

New features include:
✔️ Better stat treatments
✔️ Efficiency gains
✔️ More measurements & SM predictions
🧵

Lmao 🤣 savages

Organic Maps (@organicmaps@mastodon.social) Attached: 1 image We have a temporary glitch with GitHub—probably some contributor was geolocated in a sanctioned region (no details yet). All required documents to unlock the account have been uploa...

Absolutely appalled by GitHub not reversing the decision to lock down @organicmaps.bsky.social repo because a collaborator was in a US-sanctioned country. 😱 #opensource projects should be on high alert and consider moving away from GitHub. 🤬 mastodon.social/@organicmaps...

A measurement of the high-mass $τ\barτ$ production cross-section at $\sqrt{s}=13$ TeV with the ATLAS detector and constraints on new particles and couplings The production cross-section of high-mass $τ$-lepton pairs is measured as a function of the dilepton visible invariant mass, using 140 fb$^{-1}$ of $\sqrt{s}=13$ TeV proton-proton collision data recor...

Really happy to see this paper out: arxiv.org/abs/2503.19836
More on why, WTF it means etc later :-)

🗓️ Early-bird registration for the MCnet Summer School closes 31st March!
If you’re planning to attend, now’s the time to sign up.

Hope to see many of you in Stirling this June! 🏰⚛️🚀
🔗 indico.cern.ch/e/mcnet2025

A great chance to shape the future of collider physics simulations! Apply here: www.smartrecruiters.com/CERN/7440000...

#HEP #ParticlePhysics #PhDOpportunity #MonteCarlo #HPC #GPUComputing #LHC

Collaborate with CERN, UCL & Fermilab, with training in HEP computing & hardware acceleration. Ideal for candidates with HEP theory, computational physics, or software engineering experience.

The project develops parallelised, stable algorithms for one-loop amplitudes, optimised for GPU & HPC architectures. Work within Sherpa/Pepper to speed up Monte Carlo simulations for the HL-LHC.

The HL-LHC will generate an unprecedented volume of data, requiring faster, more precise Monte Carlo simulations. This PhD project focuses on GPU-accelerated NLO event generation to push simulations beyond current limits. Interested?

🚀 PhD Opportunity: Accelerating One-Loop Amplitude Calculations for the HL-LHC 🧵

Contur 3.1.0 · Cedar / Contur · GitLab Contur release intended for use with Rivet 4.0.3, Yoda 2.0.3, Spey 0.2.0 Includes new measurements, new statistics functionality from Spey (fitting of signal strength), and various...

🚀 Contur 3.1.0 is here! 🎉

This release is designed for use with Rivet 4.0.3 & YODA 2.0.3 and brings:
✔️ Signal strength fitting via Spey
✔️ Several new measurements
✔️ Various improvements & fixes

Get the latest version here: gitlab.com/hepcedar/con...

#HEP #OpenScience #Contur

#FOSDEM25 was an incredible mix of great talks, fascinating discussions, and the usual controlled chaos of 8,000+ FOSS enthusiasts in one place! 🎤💻 I’ve written up my reflections on the experience—check out my blog post here: blogs.ucl.ac.uk/research-sof... #FOSDEM #OpenScience #FOSS

🚀 New patch releases! We’re pleased to announce Rivet 4.0.3 and YODA 2.0.3, featuring important enhancements and bug fixes. Both are available now—get them at:
🔗 rivet.hepforge.org/
🔗 yoda.hepforge.org/

Upgrade today and let us know if you have any feedback! #HEP #OpenScience #Rivet #YODA

Want to learn more? Watch the talk here: fosdem.org/2025/schedul... and check out Rivet at rivet.hepforge.org. Happy to answer any questions—let’s keep the conversation going! #HEP #OpenResearch #OpenScience #FOSDEM

I also discussed recent Rivet developments: improved analysis capabilities, better usability, and ongoing efforts to integrate more experimental results. There’s plenty of room for contributions—get involved!

Rivet enables Monte Carlo event generators to be validated against real collider data, ensuring that our theoretical models keep pace with experimental discoveries. It’s a key tool for precision physics & BSM searches!

#LHC experiments generate petabytes of data, but without proper preservation, key analyses can become unusable over time. Rivet helps by providing a framework to encode and compare experimental results with theory reliably.

Had a great time presenting #LHC analysis preservation with Rivet at #FOSDEM (despite getting interrupted by Trump 👀)! If you missed it, the recording is now available here: fosdem.org/2025/schedul... Quick recap of what I covered—🧵

🚀 Exciting #GSoC2025 Opportunities with #MCnet!
This summer, we’re inviting passionate students to help build cutting-edge tools for particle physics research and outreach. Let’s dive in!

Want to learn more? Watch the talk here: fosdem.org/2025/schedule/event/fosdem-2025-4809-efficient-histogramming-for-high-performance-computing-in-c-with-yoda/ and check out YODA at yoda.hepforge.org. Let’s discuss high-performance data analysis! #HPC #C++ #HEP #BigData #OpenScience #FOSDEM

Recent developments include performance tuning for #HPC workflows and an HDF5-based output format. Plenty of room for contributions—interested in optimising scientific computing? Let’s talk!

YODA is built for efficiency:
• Updatable statistics, avoiding full dataset storage
• Optimised C++ data structures
• Support for event-weighted data
• Easy integration with #HEP analysis tools like Rivet

Particle physics datasets are massive. Histogramming itself isn’t a bottleneck, but inefficient implementations can slow down analyses significantly. YODA provides a lightweight, parallel-friendly, and cache-efficient solution.

My talk on HPC-friendly histogramming in C++ with YODA at #FOSDEM is now online in case you missed it: fosdem.org/2025/schedule/event/fosdem-2025-4809-efficient-histogramming-for-high-performance-computing-in-c-with-yoda/ Quick recap of key points—🧵

Diagram of YouGov results. See breakdown here: https://yougov.co.uk/politics/articles/51504-do-britons-support-shifting-to-proportional-representation

All UK parties (except Conservatives, marginally) have voter bases that prefer proportional representation over first-past-the-post.