Description
What are we working on - and what should we be working on?
I'm going to present, Pairton, my latest work on reconstruction of full hadronic ttbar events.
As part of the RESONEX program — a planned multi-lens narrow-resonance search across ~1,500 invariant mass histograms on ATLAS Run-3 data — we are exploring how to build a single, lens-agnostic background estimation framework that could scale to this many channels without per-channel hand-tuning.
The current thinking is a two-tier, fully data-driven approach: Gaussian Process Regression...
LLM-based agents are rapidly entering HEP workflows — automating analysis pipelines, assisting experimental design, and accelerating theory-experiment comparison. Yet each new project solves the same underlying problem in isolation: teaching the agent what a HEP analysis actually is. What sequence of steps takes a physics goal to a publishable result? Which control and validation regions...
Running a real AD analysis in ATLAS teaches you quickly how many things can go wrong: data reuse creates spurious excesses, false positive rates never quite reach the required threshold, and any pipeline imperfection gets amplified into a fake signal. As the field considers agentic AI for broader automated searches, these problems do not disappear — they accelerate. I want to discuss what...
We investigate whether HEP event representations, despite being trained on disparate tasks and architectures, converge to a common latent geometry — a universal event space whose structure reflects the physical content of collisions rather than the specifics of any single model. Building on recent work suggesting platonic convergence in vision and language embeddings, we develop unsupervised...
In this project I am exploring optimal detector design for the FASERν upgrade. For different detector designs, classifiers are trained and evaluated, and a surrogate model learns the relation between detector parameters and classifier performance. This surrogate is then used to suggest new detector designs for the next optimization step.
The main challenge is the trade-off between...
