Sebastian Neumayer
Chemnitz University of Technology · Germany
Inverse-problems researcher at TU Chemnitz: learned regularization for 4D Flow CMR and fast Gaussian kernel sums.
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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—
The snapshot reports h-index 0 and i10-index 0 against 8 works in the recent three-year window, suggesting sparse indexing of an early record rather than a measured output level. The output axis would aggregate total and per-year publication counts across indexed venues; no score is computed yet.
Impact—
No impact score is available: the record lists h-index 0 and i10-index 0 with no per-year citation counts, so citation aggregation appears incomplete for this author identity. The impact axis would measure citation accumulation, h/i10-based influence, and per-paper reach across indexed work.
Focus0/100
The focus score is 0, reflecting a perfectly even OpenAlex field distribution: medical image reconstruction / computational inverse problems and machine learning — kernel methods and numerical analysis each account for 50%. Topic counts confirm the spread — 3 works each in mathematical analysis and transform methods and in mathematical approximation and integration, and 2 each in medical imaging, denoising, and numerical methods in inverse problems.
Collaboration22/100
Collaboration scores 22, computed from a snapshot of only 2 distinct coauthors, with German Shâma Wache and Nicolaj Rux the most frequent. Both reviewed preprints carry short author lists and place Neumayer in the last author position, non-corresponding — a pattern consistent with compact mathematics-style teams rather than large lab or clinical consortia.
Activity64/100
Activity scores 64, grounded in recent3yWorks of 8 and a countsByYear entry recording 8 outputs in 2026 alone. Both Evolve-reviewed preprints date to 2026, so the visible record is concentrated in a single, currently active publishing year, with a steady clip despite thin long-run indexing.
Reproducibility55/100
Reproducibility averages 55 across the 2 Evolve-reviewed papers, both rated medium readiness. CLEAR reports an 8.3k-parameter regularizer beating LLR and FlowVN at 10–50x speedup, but with few baselines and no variance reporting; the Gauss-sums paper shows real fp16 speedups while its abstract slightly overstates accuracy and memory wins.
Research summary
Sebastian Neumayer is a researcher at Chemnitz University of Technology (Germany) whose record, resolved from OpenAlex bibliographic data, sits at the interface of computational inverse problems and the numerical analysis of machine learning methods. The Evolve catalog reviews two 2026 arXiv preprints on which he appears in the last author position: CLEAR, an 8.3k-parameter learned explicit analytical regularizer for ultra-accelerated 4D Flow CMR reconstruction that beats LLR and FlowVN on CMRx4DFlow2026 metrics at 10–50x speedup, and Fast Gauss Sums via Flash Attention, which computes Gaussian kernel sums through flash attention with real fp16 speedups for kernel methods such as maximum mean discrepancy and Stein variational gradient descent. His OpenAlex field distribution splits evenly (50/50) between medical image reconstruction / computational inverse problems and machine learning — kernel methods and numerical analysis, with leading topics in mathematical analysis and transform methods, mathematical approximation and integration, medical imaging, denoising, and numerical methods in inverse problems. The snapshot records 8 works in the recent three-year window, 2 distinct coauthors, and h-index 0 / i10-index 0, a pattern consistent with thin indexing of an active record rather than an established citation history.
Collaboration pattern
Collaboration is compact: the snapshot lists 2 distinct coauthors (collaboration score 22), with German Shâma Wache and Nicolaj Rux as the most frequent. Both reviewed preprints carry short author lists and place Neumayer in the last author position, non-corresponding, consistent with small mathematics-style teams; no large consortium or clinical-group authorship appears in the reviewed record.
Research keywords
Highlights
8.3k-parameter learned regularizer for ultra-accelerated 4D Flow CMR reconstruction
Gaussian kernel sums computed through flash attention
Timeline
- 2026CLEAR: Complex Learned Explicit Analytical Regularization for Ultra-Accelerated 4D Flow CMR Reconstruction appears on arXiv (medical image reconstruction / computational inverse problems).
- 2026Fast Gauss Sums via Flash Attention appears on arXiv, connecting Gaussian kernel sums to flash attention (machine learning — kernel methods and numerical analysis).
- 2026The OpenAlex snapshot records 8 works in the recent three-year window with 2 distinct coauthors, while h-index and i10-index both read 0.
Papers in the Evolve catalog
Frequent co-authors
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.
- Bibliometrics come from OpenAlex only; no ORCID is linked, so the record is name-resolved from public bibliographic data and may merge or miss works.
- The snapshot shows h-index 0 and i10-index 0 with countsByYear covering only 2026, so output and impact dimensions are unscored and long-run trends cannot be assessed.
- Evolve review coverage is limited to 2 arXiv preprints (2026), not peer-reviewed publications; conclusions may shift with journal versions.
- The reproducibility mean (55) rests on only 2 reviewed papers and should be read as an early-stage signal.