Xu Yuqing
—
Early-career applied ML researcher focused on photovoltaic power forecasting with transparent ensemble benchmarks.
2
papers here
2
reviewed by Evolve
1
first-author
0
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: Evolve holds no verified publication counts for this researcher — only the two reviewed 2026 arXiv preprints — so total output, yearly productivity, and venue spread cannot be measured until fuller bibliometric records are linked.
Impact—
Not yet scored: the input data contains no citation counts, h-index, or uptake signals, so this axis — which would track how often the two PV forecasting papers are cited and used downstream — stays unassessed rather than estimated.
Focus0/100
The radar records a focus score of 0/100. The underlying field distribution splits evenly (0.5/0.5) between two near-identical labels — 'applied machine learning for renewable energy forecasting' and 'applied machine learning / renewable energy forecasting' — so both catalog papers sit in one narrow applied-ML/PV-forecasting area. With only two papers, this concentration signal is thin, and the zero score likely reflects the sparse record rather than scattered research interests.
Collaboration32/100
Collaboration scores 32/100, based on a small recorded network: 4 distinct coauthors — Zhou Liguo, Sun Ze, Yu Lei, and Jiang Mingming — span both 2026 arXiv papers. Xu is first author on 'Horizon-specific Expert Fusion' and a middle author on 'Ensemble Complexity in Photovoltaic Forecasting,' with no corresponding-author role recorded. The pattern suggests a compact, stable team producing preprints together rather than a wide cross-institution network.
Activity—
Not yet scored: no yearly publication-volume series is available — the catalog shows only two preprints, both dated 2026 — so publication cadence and recent-productivity trends cannot be computed for this axis.
Reproducibility55/100
Reproducibility scores 55/100, a mean taken over the 2 papers Evolve has reviewed, each rated 'medium' readiness. Both papers evaluate on public benchmarks — GEFCom2014 and OPSD — and report component ablations plus explicit significance thresholds (an own 0.2-percentage-point improvement bar), practices that support independent replication. Limits remain: headline gains are site-specific, and the oracle-weather framing restricts broader conclusions.
Research summary
Xu Yuqing is an early-stage applied machine learning researcher working on photovoltaic (PV) power forecasting, with a catalog footprint built entirely on public benchmark datasets. The Evolve catalog holds two 2026 arXiv papers, both co-authored with the same four recurring collaborators — Zhou Liguo, Sun Ze, Yu Lei, and Jiang Mingming. The first-authored paper, 'Horizon-specific Expert Fusion for Photovoltaic Power Forecasting,' builds an eight-expert hierarchical ensemble that edges LightGBM by 4.11% at one station while openly failing its own 0.2-percentage-point significance bar; Evolve judges it rigorous and honest, with small, site-specific gains. The companion paper, 'Ensemble Complexity in Photovoltaic Forecasting,' audits ensemble components and finds fusion beats matched boosting only on OPSD after correction, that weather gating adds nothing, and that oracle-weather framing limits broader claims. Both papers were rated 'medium' for reproducibility readiness, giving a mean reproducibility score of 55/100 across the 2 assessed papers. The radar records a collaboration score of 32/100 over 4 distinct coauthors and a focus score of 0 even though the field distribution splits evenly between two near-identical labels of one applied-ML/renewable-energy-forecasting area. Output, impact, and activity remain unscored because no publication-volume or citation data is available; this is a name-resolved, early-stage profile.
Collaboration pattern
Xu Yuqing works within a compact, recurring team: the two 2026 arXiv papers share four distinct coauthors — Zhou Liguo, Sun Ze, Yu Lei, and Jiang Mingming. Xu leads one paper as first author and sits in the middle of the author list on the other, with no corresponding-author record. The collaboration score of 32/100 reflects this small, stable footprint rather than a broad cross-institution network; all catalog output is preprint-stage applied ML for PV forecasting.
Research keywords
Highlights
Eight-expert fusion edges LightGBM by 4.11% — but fails its own bar
Component audit: ensemble complexity only pays on one corrected dataset
Medium reproducibility on public benchmarks
Both catalog papers are rated 'medium' for reproducibility readiness, giving a mean score of 55/100; evaluation on public datasets (GEFCom2014, OPSD) with reported ablations makes the work checkable, though site-specific gains temper direct reuse.
Timeline
- 2026First-author preprint: 'Horizon-specific Expert Fusion for Photovoltaic Power Forecasting' appears on arXiv, testing an 8-expert hierarchical ensemble against LightGBM.
- 2026Co-authors 'Ensemble Complexity in Photovoltaic Forecasting,' an audit of fusion, matched boosting, and weather gating on GEFCom2014 and OPSD.
- 2026A recurring team forms: the same four coauthors (Zhou Liguo, Sun Ze, Yu Lei, Jiang Mingming) span both papers, which Evolve rates at medium reproducibility readiness (mean 55/100).
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.
- Bibliometric signals come from public bibliographic data only; no ORCID, institution, or country is verified, so this profile is name-resolved and may merge or split same-name authors.
- Only two arXiv preprints, both from 2026, have been reviewed by Evolve; no journal articles, citation counts, or h-index are available, leaving output, impact, and activity unscored.
- Coauthor names are listed in pinyin as provided; Chinese character renderings (including that of the profile subject) are not confirmed.
- Reproducibility and collaboration scores cover only the two reviewed papers and may change as the catalog grows.