Speaker
Thanush Sivagnanalingam
(University Heidelberg)
Description
Neural network training for LHC event generation should, ideally, benefit from common high-level patterns in different processes. We propose novel conditioning schemes for continuous parameters, process labels, and Feynman diagrams. We employ pre-trained LLMs as multi-modal foundation models to provide descriptive embeddings for an autoregressive transformer. With such high-level physics-inductive bias the generative networks converge faster, provide better result, and generalize to unseen processes.
Authors
Daniel Schiller
(Institute for Theoretical Physics, Heidelberg University)
Henning Bahl
Thanush Sivagnanalingam
(University Heidelberg)
Tilman Plehn
(Institut für Theoretische Physik, Universität Heidelberg; Interdisciplinary Center for Scientific Computing (IWR), Universität Heidelberg)