Preprint

PUBDB-2026-01685

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png SPADE: Split-and-Delay Embeddings for Autoregressive High-Granularity Calorimeter Simulation

Birk, J.Extern* ; Gaede, F.DESY* ; Hallin, A.Extern* ; Kasieczka, G.Extern* ; Mozzanica, M.Extern* ; Rose, H. (Corresponding author)Extern*

2026

This record in other databases:    

Please use a persistent id in citations: doi:10.3204/PUBDB-2026-01685

Report No.: arXiv:2606.11304

Abstract: We introduce SPADE (SPlit And Delay Embeddings), an autoregressive transformer for sequences whose tokens carry multiple features. Rather than embedding these features jointly, SPADE embeds them independently. Delaying each feature stream relative to the previous one allows intra-token correlations to be learned by the standard self-attention mechanism. Applied to point-cloud calorimeter shower generation in the highly granular ILD detector, SPADE is competitive with the state of the art AllShowers model on photon showers, and substantially outperforms its VQ-VAE-based predecessor OmniJet-$α_C$. The mechanism is applicable to any generative task with multi-feature tokens, enabling LLM-style pretraining workflows for higher-dimensional data.

Note: 20 pages, 13 figures SciFM consortium (05D25GU4)

---

Contributing Institute(s):

1. Technol. zukünft. Teilchenph. Experim. (FTX)

Research Program(s):

1. 623 - Data Management and Analysis (POF4-623) (POF4-623)
2. DFG project G:(GEPRIS)390833306 - EXC 2121: Das Quantisierte Universum II (390833306) (390833306)
3. DFG project G:(GEPRIS)460248186 - PUNCH4NFDI - Teilchen, Universum, Kerne und Hadronen für die NFDI (460248186) (460248186)

Experiment(s):

1. Future Circular Collider

Appears in the scientific report 2026

---

Database coverage:
;

---