Grounding X-Ray Simulations in Reality: Physics-Aware Calibration for X-Ray Models
Bibliographic record
Abstract
The demand for larger datasets is growing as AI and machine learning become vital in signal processing for high-rate experiments. Yet, such sizable datasets are difficult to obtain through real-world setups, prompting the use of Monte Carlo simulation frameworks, such as Geant4, to generate synthetic data. However, accurately reproducing experiments through simulation poses a significant challenge: key parameters – including material properties, source distributions, and detector geometry – can be poorly documented or unknown. In practice, aligning simulation outputs with real-world measurements is time-consuming and often relies on expert intuition or brute-force parameter sweeps. To address this, we propose a physics-aware optimization approach that automatically calibrates simulation parameters to match experiments. Treating Geant4 as a black box, our method leverages derivative-free optimization techniques to minimize a data-driven loss between simulated and observed outputs. Physics-informed constraints are incorporated through clipping and penalty-based regularization to enforce plausible solutions. We evaluate our approach on synthetic X-ray imaging data, comparing recovered parameters to the known ground truth and a real-world dataset with unknown parameters. In both cases, we assess image quality using standard SSIM, demonstrating accurate configuration recovery while substantially reducing manual calibration effort. Our method generalizes to various experimental setups where simulation fidelity is essential but full system specification is unavailable.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.011 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".