Researchers have introduced the SUrrogate ReFerence (SURF) method to validate generative models in high-energy physics, specifically for jet tagging. This framework allows for exact Neyman-Pearson tests by training a target model on samples from a tractable surrogate, which is itself trained on real data. The study suggests that current jet taggers may be approaching their statistical limits, while autoregressive GPT models appear to overestimate separation power, providing a misleading view of these fundamental limits. AI
IMPACT Introduces a novel validation technique for generative models, potentially improving the accuracy of scientific simulations and analyses.
RANK_REASON The cluster contains a research paper detailing a new method for validating generative models in a specific scientific domain. [lever_c_demoted from research: ic=1 ai=1.0]
- alphaXiv
- arXiv
- DagsHub
- EPiC-FM
- generative pre-trained transformer
- Hugging Face
- Ian Pangilinan
- IArxiv
- JetClass
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →