Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.21035 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908987879849984 |
|---|---|
| author | Visive, Ambre de Austri, Roberto Ruiz Moskvitina, Polina Nellist, Clara Caron, Sascha |
| author_facet | Visive, Ambre de Austri, Roberto Ruiz Moskvitina, Polina Nellist, Clara Caron, Sascha |
| contents | Anomaly detection in High Energy Physics requires identifying rare signals against overwhelming backgrounds, without prior knowledge of the signal. We present the first application of masked-token prediction, a technique from Large Language Models, to this problem. A lightweight encoder architecture trained solely on background events captures the structure of Standard Model (SM) physics; at inference, sequences deviating from this learned structure are flagged as anomalous. We evaluate the approach on searches for four-top-quark production and supersymmetric gluino pair production, both featuring top-rich final states with substantial missing transverse energy, covering SM and beyond the Standard Model (BSM) scenarios. Strong performance on the four-top signature, which closely resembles background, demonstrates the method's sensitivity to subtle deviations. We further show that the tokenization strategy significantly impacts performance: deep-learned tokenization via vector-quantized variational autoencoders (VQ-VAE) outperforms look-up table tokenization. Comparison with established anomaly detection baselines confirms robustness. These results highlight the potential of token-based collider data representations combined with transformer architectures for new-physics discovery. Once trained on SM background, the model transfers across different BSM searches, enabling scalable, model-independent anomaly detection at reduced computational cost. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_21035 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Masked-Token Prediction for Anomaly Detection at the Large Hadron Collider Visive, Ambre de Austri, Roberto Ruiz Moskvitina, Polina Nellist, Clara Caron, Sascha High Energy Physics - Phenomenology High Energy Physics - Experiment Anomaly detection in High Energy Physics requires identifying rare signals against overwhelming backgrounds, without prior knowledge of the signal. We present the first application of masked-token prediction, a technique from Large Language Models, to this problem. A lightweight encoder architecture trained solely on background events captures the structure of Standard Model (SM) physics; at inference, sequences deviating from this learned structure are flagged as anomalous. We evaluate the approach on searches for four-top-quark production and supersymmetric gluino pair production, both featuring top-rich final states with substantial missing transverse energy, covering SM and beyond the Standard Model (BSM) scenarios. Strong performance on the four-top signature, which closely resembles background, demonstrates the method's sensitivity to subtle deviations. We further show that the tokenization strategy significantly impacts performance: deep-learned tokenization via vector-quantized variational autoencoders (VQ-VAE) outperforms look-up table tokenization. Comparison with established anomaly detection baselines confirms robustness. These results highlight the potential of token-based collider data representations combined with transformer architectures for new-physics discovery. Once trained on SM background, the model transfers across different BSM searches, enabling scalable, model-independent anomaly detection at reduced computational cost. |
| title | Masked-Token Prediction for Anomaly Detection at the Large Hadron Collider |
| topic | High Energy Physics - Phenomenology High Energy Physics - Experiment |
| url | https://arxiv.org/abs/2604.21035 |