Related papers: GLOW: A Unified Particle Flow Transformer
The ALICE Collaboration at CERN developed a 3D visualisation tool capable of displaying a representation of collected collision data (particle trajectories, clusters and calorimeter towers) called the Event Display. The Event Display is…
The studies presented in this paper provide a first experimental test of the Particle Flow Algorithm (PFA) concept using data recorded in high granularity calorimeters. Pairs of overlaid pion showers from CALICE 2007 test beam data are…
3D world models (i.e., learning-based 3D dynamics models) offer a promising approach to generalizable robotic manipulation by capturing the underlying physics of environment evolution conditioned on robot actions. However, existing 3D world…
Sampling useful three-dimensional molecular structures along with their most favorable conformations is a key challenge in drug discovery. Current state-of-the-art 3D de-novo design flow matching or diffusion-based models are limited to…
Unified and scalable Transformers have recently achieved remarkable success in modeling diverse phenomena traditionally associated with computer graphics, such as 3D visual effects, rendering processes, and motion in videos. In this work,…
We demonstrate transfer learning capabilities in a machine-learned algorithm trained for particle-flow reconstruction in high energy particle colliders. This paper presents a cross-detector fine-tuning study, where we initially pretrain the…
The aim of this paper is to provide a mathematical analysis of transformer architectures using a self-attention mechanism with layer normalization. In particular, observed patterns in such architectures resembling either clusters or uniform…
Future collider experiments require unprecedented precision in measurements of Higgs, electroweak, and flavour observables, placing stringent demands on event reconstruction. The achievable precision on Higgs couplings scales directly with…
This paper introduces DynaFlow, a novel framework that embeds a differentiable simulator directly into a flow matching model. By generating trajectories in the action space and mapping them to dynamically feasible state trajectories via the…
Learning-based optical flow estimation has been dominated with the pipeline of cost volume with convolutions for flow regression, which is inherently limited to local correlations and thus is hard to address the long-standing challenge of…
In this talk we discuss several interesting aspects of collective flow in small systems in reference to planned future experiments. First, we bring in the novel possibility of exploring elliptic flow with heavy nuclei collisions on…
Normalizing flows model complex probability distributions using maps obtained by composing invertible layers. Special linear layers such as masked and 1x1 convolutions play a key role in existing architectures because they increase…
Simulating particle detector response is the single most expensive step in the Large Hadron Collider computational pipeline. Recently it was shown that normalizing flows can accelerate this process while achieving unprecedented levels of…
One of the most important aspects of detector development for the ILC is a good jet energy resolution sigma_E/E. To achieve the goal of high precision measurements sigma_E/E = 0.30/sqrt(E(GeV)} is proposed. The particle flow approach…
Recently, diffusion-based video generation models have achieved significant success. However, existing models often suffer from issues like weak consistency and declining image quality over time. To overcome these challenges, inspired by…
Electron beam lithography (EBL) is a promising maskless solution for the technology beyond 14nm logic node. To overcome its throughput limitation, recently the traditional EBL system is extended into MCC system. %to further improve the…
We present a novel, integrated control framework designed to achieve seamless transitions among a spectrum of inverter operation modes. The operation spectrum includes grid-forming (GFM), grid-following (GFL), static synchronous compensator…
6D object pose estimation is crucial for robotic perception and precise manipulation. Occlusion and incomplete object visibility are common challenges in this task, but existing pose refinement methods often struggle to handle these issues…
Algorithms based on the particle flow approach are becoming increasingly utilized in collider experiments due to their superior jet energy and missing energy resolution compared to the traditional calorimeter-based measurements. Such…
Generative modeling has emerged as a powerful paradigm for representation learning, but its direct applicability to challenging fields like medical imaging remains limited: mere generation, without task alignment, fails to provide a robust…