Benchmarking Sensor Robustness in Plasma Diagnostic Models: A Systematic Evaluation on TokaMark

Hugging Face Daily Papers Papers
plasma-diagnostics tokamak robustness benchmarking machine-learning sensor-failure

Summary

This paper introduces the first systematic robustness benchmark for plasma diagnostic machine learning models using the TokaMark dataset, evaluating architectures across six sensor failure scenarios and introducing a Robustness Score for cross-architecture comparison.

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.
Original Article
View Cached Full Text

Cached at: 07/20/26, 09:26 PM

Paper page - Benchmarking Sensor Robustness in Plasma Diagnostic Models: A Systematic Evaluation on TokaMark

Source: https://huggingface.co/papers/2607.11915

Abstract

Plasmadiagnosticmodelsfortokamakfusiondevicesarealmostuniversallyevaluatedonclean,completesensordata.Inpractice,fusiondiagnosticsfailregularly:acquisitionsystemsstartlate,individualsensorsdie,andsignaldropoutsclusterpreciselywhenaplasmadisruptionisapproaching.WepresentthefirstsystematicrobustnessbenchmarkforplasmadiagnosticMLusingtheTokaMarkdatasetof11,573MASTshots,evaluatingXGBoost,LSTM,Transformer,andtheTokaMarkCNNbaselineacrosssixphysically-groundedfailurescenariosandthreeimputationstrategies.WeintroducetheRobustnessScore(RS)forstandardizedcross-architecturecomparison.Ourcentralfindingisthatdisruption-proximatesensorfailure(corruptioninjectedinthefinalwindowtimesteps)collapsessequencemodelperformance(LSTM+212%NRMSE)whileastatisticalfeaturemodelremainscomparativelystable(XGBoost+37%).Forward-fillimputationeliminatesnearlyalldegradationfromrandomdropoutforsequencemodels(LSTM+57%to~0%),butofferslittlehelpwhentheendofthewindowiscorrupted.Shot-levelalarmevaluationusingground-truthdisruptiontimestampsrevealsthatLSTMalarmdetectioncollapsestoTPR=0.00underproximatesensorfailure,whilemean-fillimputationrecoversittoTPR=1.00,areversalofthepatternobservedinNRMSE.Plasmacurrentemergesasthesinglemostcriticaldiagnosticacrossallarchitectures(+73%to+140%uponremoval).Code,data,andtrainedcheckpointsareavailableathttps://github.com/Neerav-Gupta/tokamark-robustness.

View arXiv pageView PDFGitHub3Add to collection

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 paper0

No model linking this paper

Cite arxiv.org/abs/2607.11915 in a model README.md to link it from this page.

Datasets citing this paper1

#### Neerav-Gupta/tokamark-robustness-data

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.11915 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

TS-Fault: Benchmarking Time Series Forecasters Against Structural Faults

arXiv cs.LG

This paper introduces TS-Fault, a benchmark for evaluating time series forecasting models under structured fault scenarios like broken dependencies and regime changes, finding that clean-data accuracy often anti-correlates with robustness and that foundation models are especially fragile.

Benchmarking Large Language Models on Multi-Sensor Physical Hazard Assessment

arXiv cs.AI

The paper benchmarks five large language models on multi-sensor physical hazard assessment, revealing that all tested models fail to produce precautionary warnings when multiple sensors are simultaneously elevated below their individual safety limits, while achieving near-perfect accuracy on single-sensor violations.

Unsteady Metrics and Benchmarking Cultures of AI Model Builders

arXiv cs.AI

This paper introduces Benchmarking-Cultures-25, a dataset analyzing how AI model builders selectively highlight benchmarks in press releases. It finds a fragmented evaluation landscape with limited cross-model comparability, arguing that benchmarks are used as narrative devices for market positioning rather than standardized scientific measurement.