Tariq Benamara

Cenaero (Belgium)

Unclaimed · AI-structured profileORCID OpenAlex

Scientific-ML researcher at Cenaero applying neural operators and neural fields to aerodynamic surrogate modeling

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

—

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-15) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.

Research profile

Output—

Not yet scored: this snapshot contains no verified publication-count or per-year output totals for Tariq Benamara. The output axis would measure cumulative and annual production volume across indexed venues once a fuller bibliographic record is available.

Impact—

Not yet scored: citation counts, citation-based indices, and downstream usage are absent from the available data. The impact axis would measure citations, adoption, and influence of this researcher's work on the aerodynamics and scientific-ML communities.

Focus0/100

The radar focus score is 0/100, reflecting the observed field distribution: output is split evenly (0.5/0.5) between scientific machine learning for aerodynamics (CFD surrogate modeling) and scientific machine learning / turbomachinery aerodynamics. In practice the two strands are adjacent — both target neural surrogates for aerodynamic prediction — but the even split across two labeled fields leaves no single dominant concentration under the scoring method.

Collaboration41/100

Collaboration scores 41/100 on the radar, built on 7 distinct coauthors in the snapshot. Frequent collaborators — Lionel Salesses, Caroline Sainvitu, Joachim Dominique, Théo Flament and Franck Mastrippolito — suggest a stable industrial research group at Cenaero rather than a wide academic network; both catalog papers are team-authored with Benamara in middle or last (non-corresponding) positions.

Activity—

Not yet scored: the snapshot lacks timestamped publication events, so recent publication rate and year-over-year cadence cannot be computed. The activity axis would measure how consistently and recently this researcher publishes, using dated records across indexed sources.

Reproducibility55/100

Reproducibility scores 55/100 (reproMean 55 across 2 assessed papers), with both catalog entries rated medium readiness. The ONERA CRM challenge paper reports a stepwise ablation, while the rotor-blade p-FNM study evaluates against an internal VAE-RNN baseline on a single geometry without external baselines — consistent with solid but not fully open, externally validated methodology.

Research summary

Tariq Benamara is a researcher at Cenaero (Belgium) working at the interface of scientific machine learning and computational aerodynamics. Evolve's reviewed catalog contains two of his 2026 arXiv preprints, both in team author positions (middle on one, last on the other, neither corresponding). The first, 'Periodic Neural Mapping for Unsteady Rotor-Blade Pressure and Aeroelastic Load Prediction', introduces p-FNM, which adds periodic time encoding to the Fourier Neural Operator and reduces pressure and generalized-aerodynamic-force errors relative to an internal VAE-RNN baseline; Evolve flags a single-geometry evaluation and the absence of external baselines. The second, 'Neural Field Ensembles for Aerodynamic Surface Prediction', was the winning solution to the ONERA CRM Wall Distribution 2025 Challenge, scoring 8.81 against an 8.64 baseline, with solid stepwise ablation though leaderboard-driven selection. Radar profiles show a 50/50 field split between scientific machine learning for aerodynamics (CFD surrogate modeling) and scientific machine learning / turbomachinery aerodynamics, a collaboration score of 41/100 built on 7 distinct coauthors, and a reproducibility score of 55/100 with both papers at medium readiness. Frequent coauthors — Lionel Salesses, Caroline Sainvitu, Joachim Dominique, Théo Flament and Franck Mastrippolito — point to a stable Cenaero research group. Output, impact and activity axes are not yet scored, and no citation metrics are available in this snapshot.

Collaboration pattern

Collaboration at Cenaero appears group-centric: 7 distinct coauthors in the radar snapshot yield a 41/100 collaboration score, and the recurring names — Lionel Salesses, Caroline Sainvitu, Joachim Dominique, Théo Flament and Franck Mastrippolito — appear across the 2026 catalog papers. Benamara sits in middle or last author positions on team-authored arXiv preprints, consistent with applied, project-driven industrial research rather than solo or wide-network academic publishing.

Research keywords

Scientific machine learningFourier Neural OperatorNeural fieldsImplicit neural representationsReduced-order modelingAerodynamic surrogate modelingCFDUnsteady turbomachinery aerodynamicsGeneralized aerodynamic forcesAeroelasticityFourier feature encoding

Highlights

Winning solution to the ONERA CRM Wall Distribution 2025 Challenge
A conditional neural-field ensemble for aerodynamic surface prediction scored 8.81 versus the 8.64 baseline; Evolve's review notes solid stepwise ablation, though model selection was leaderboard-driven.
Periodic time encoding for unsteady rotor-blade prediction (p-FNM)
Extends the Fourier Neural Operator with periodic time mapping to predict unsteady rotor-blade pressure and generalized aerodynamic forces, cutting pressure and GAF errors against an internal VAE-RNN baseline; demonstrated on a single geometry.
Linking neural surrogates to aeroelastic load estimation
The p-FNM mapping targets not only unsteady surface pressure but also generalized aerodynamic forces, connecting neural operator surrogates to aeroelastic load prediction for turbomachinery rotors.

Timeline

2026Co-authored (middle position) 'Periodic Neural Mapping for Unsteady Rotor-Blade Pressure and Aeroelastic Load Prediction', introducing p-FNM periodic time encoding for FNO-based pressure and generalized-aerodynamic-force prediction, released on arXiv.
2026Co-authored (last position) 'Neural Field Ensembles for Aerodynamic Surface Prediction', the winning solution to the ONERA CRM Wall Distribution 2025 Challenge, released on arXiv.
2026Both reviewed arXiv preprints appeared in 2026, forming the current evidence base for this full profile in scientific machine learning for aerodynamics.

Papers in the Evolve catalog

Frequent co-authors

Lionel Salesses×2 Caroline Sainvitu×1 Joachim Dominique×1 Théo Flament×1 Franck Mastrippolito×1 Lionel Salesses×1 Tariq Benamara×1

About this profile

This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.

Radar scores and bibliometrics are computed from a limited OpenAlex-based snapshot; output, impact and activity axes remain unscored.
Only two 2026 arXiv preprints are in the reviewed catalog; no citation counts, h-index or journal publication record were available, so none are quoted.
Profile is name-resolved from public bibliographic data via ORCID 0000-0001-5342-2408; country is not specified in the source data.
Both catalog papers carry team author positions (middle/last, non-corresponding), so individual contribution levels are not independently verified.