Benchmarking Sensor Robustness in Plasma Diagnostic Models: A Systematic Evaluation on TokaMark
Abstract
Plasma diagnostic models for tokamak fusion devices are almost universally evaluated on clean, complete sensor data. In practice, fusion diagnostics fail regularly: acquisition systems start late, individual sensors die, and signal dropouts cluster precisely when a plasma disruption is approaching. We present the first systematic robustness benchmark for plasma diagnostic ML using the TokaMark dataset of 11,573 MAST shots, evaluating XGBoost, LSTM, Transformer, and the TokaMark CNN baseline across six physically-grounded failure scenarios and three imputation strategies. We introduce the Robustness Score (RS) for standardized cross-architecture comparison. Our central finding is that disruption-proximate sensor failure (corruption injected in the final window timesteps) collapses sequence model performance (LSTM +212% NRMSE) while a statistical feature model remains comparatively stable (XGBoost +37%). Forward-fill imputation eliminates nearly all degradation from random dropout for sequence models (LSTM +57% to ~0%), but offers little help when the end of the window is corrupted. Shot-level alarm evaluation using ground-truth disruption timestamps reveals that LSTM alarm detection collapses to TPR=0.00 under proximate sensor failure, while mean-fill imputation recovers it to TPR=1.00, a reversal of the pattern observed in NRMSE. Plasma current emerges as the single most critical diagnostic across all architectures (+73% to +140% upon removal). Code, data, and trained checkpoints are available at https://github.com/Neerav-Gupta/tokamark-robustness.
Community
Sensor failures are common in tokamak fusion devices, and they cluster precisely when a plasma disruption is approaching, exactly when reliable prediction matters most. Robustness to this kind of failure is one of the key hurdles standing between today's plasma diagnostic ML models and safe, reliable operation in commercial fusion reactors.
I built a systematic robustness benchmark for plasma diagnostic ML on TokaMark (11,573 MAST tokamak shots), testing XGBoost, LSTM, Transformer, and a CNN baseline across 6 realistic sensor failure scenarios.
Code, data, and trained checkpoints are all open: github.com/Neerav-Gupta/tokamark-robustness
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Experiment-free disruption prediction for new devices enabled by synthetic diagnostic data augmentation (2026)
- Hierarchical Multi-to-Single-Modal Knowledge Distillation for Disruption Prediction in EAST (2026)
- Lightweight CNN-Based Anomaly Detection for High Voltage Converter Modulators in the Spallation Neutron Source (2026)
- Machine-Learning-Enabled Full-State Reconstruction of Fusion Plasmas from Minimal Sensor Measurements (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend
Get this paper in your agent:
hf papers read 2607.11915 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 1
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper