Hao Wang

—

Unclaimed · AI-structured profileOpenAlex

ML researcher spanning smart-grid analytics and language models, from smart-meter PV-EV detection to LLM benchmarking

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

—

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-09-28) 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—

Output metrics — total publications, per-year counts, and output-volume percentiles — are not available in the current bibliometric snapshot, so this axis is unscored. It would measure the scale and consistency of Wang's publication record across venues and years; only three 2026 catalog papers are evidenced.

Impact—

Impact measures — citation counts, h-index, and field-normalized citation percentiles — are absent from the current data, so this axis cannot be scored. It would quantify how widely Wang's three 2026 catalog papers and any prior work are cited by the research community.

Focus100/100

Focus scores a maximum 100: the field distribution attached to this profile contains exactly one labeled field, applied machine learning for smart grids / energy informatics, with a count of 1 — the reviewed smart-meter PV-EV paper. Within the Evolve catalog this is the only field-tagged output, so the measured focus is total, though it rests on a single labeled paper; two further 2026 arXiv papers on language-model benchmarking and role-playing agents remain unlabeled.

Collaboration64/100

Collaboration scores 64, grounded in 24 distinct coauthors in the raw bibliometric record. Twelve recur as frequent coauthors, including Yawen Ouyang, Xinbo Zhang, Ziyuan Ma, Wenjie Li, Xiaoqing Zheng, Xuefeng Yan, Lei Bai, and Ya-Qin Zhang. Across the three 2026 catalog papers Wang holds the last author position twice and the middle position once, consistent with a mix of senior and contributing roles within a moderately sized, stable network.

Activity—

Activity metrics — publication recency, per-year cadence, and venue coverage over a multi-year window — are not yet available, so this axis is unscored. It would track how steadily Wang publishes over time; the snapshot currently evidences only the three 2026 arXiv catalog papers.

Reproducibility20/100

Reproducibility scores 20 (reproMean 20), computed from one assessed paper. The reviewed smart-meter study is rated low readiness: Evolve notes the BiLSTM+DTW pipeline achieves strong detection metrics on AusNet AMI data, but the PV/EV labels are heuristic proxies rather than verified ground truth and the underlying smart-meter data stays private, limiting independent replication.

Research summary

Hao Wang (name resolved from public bibliographic records; Hao Wang is a common name) is a machine-learning researcher whose reviewed output concentrates on applied machine learning for smart grids and energy informatics, while also reaching into language-model research. The Evolve catalog holds three 2026 arXiv papers. The flagship reviewed study, Uncovering Residential PV-EV Co-Adoption from Smart-Meter Data, applies DTW k-means clustering with DBA averaging and BiLSTM classification to AusNet smart-meter data to detect residential solar and electric-vehicle adoption and derive load archetypes for demand-side planning; Evolve's verdict credits strong detection metrics but flags heuristic proxy labels and private data, yielding low reproducibility readiness (reproducibility dimension score: 20, from one assessed paper). Wang is last author on this work and on Adversarial Closed-Loop Curriculum for Evolving Role-Playing Agents, and a middle author on PFArena: Benchmarking Language Models for Protein Modification, indicating both supervisory and collaborative roles. The profile shows a maximum focus score of 100, its single labeled field being applied ML for smart grids, and a collaboration score of 64 with 24 distinct coauthors. Output, impact, and activity dimensions are not yet scored.

Collaboration pattern

The raw record lists 24 distinct coauthors; a stable core of twelve recurs, including Yawen Ouyang, Xinbo Zhang, Ziyuan Ma, Yixin Wu, Wenbin Liao, Feiran Zhang, Wenjie Li, Lihao Wang, Xiaoqing Zheng, Xuefeng Yan, Lei Bai, and Ya-Qin Zhang. Authorship positions across the 2026 catalog — last author twice, middle once — suggest a senior role within one cluster and contributory collaboration in another, consistent with the moderate collaboration score of 64.

Research keywords

smart meter dataPV-EV co-adoption detectionload profilingDTW k-means clusteringDBABiLSTMdemand-side planninglanguage model benchmarkingprotein modificationrole-playing agentscurriculum learningadversarial training

Highlights

Smart-meter detection of residential PV-EV co-adoption
Last-author work applies DTW k-means clustering with DBA averaging and BiLSTM classifiers to AusNet smart-meter data, deriving load archetypes and detecting PV/EV co-adoption for demand-side planning; Evolve credits strong detection metrics but flags heuristic proxy labels and private data (low reproducibility readiness).
Co-author on PFArena, a benchmark for protein-modification language models
A 2026 arXiv paper benchmarking language models on protein modification tasks, with Wang in a middle author position — a second application domain beyond the energy focus. Evolve has not yet reviewed this paper, so no field label, verdict, or reproducibility data is attached.
Last-author work on adversarial curricula for role-playing agents
Wang is last author of this 2026 arXiv paper on adversarial closed-loop curriculum training for evolving role-playing agents, extending the profile into LLM agent training. Evolve has not yet reviewed it, so no keywords, verdict, or reproducibility assessment exists.

Timeline

2026Published the reviewed smart-meter study on residential PV-EV co-adoption (arXiv) as last author — the profile's only field-tagged and Evolve-assessed paper.
2026Appeared as a middle author on PFArena: Benchmarking Language Models for Protein Modification (arXiv).
2026Published Adversarial Closed-Loop Curriculum for Evolving Role-Playing Agents (arXiv) as last author.

Papers in the Evolve catalog

Frequent co-authors

Yawen Ouyang×1 Xinbo Zhang×1 Ziyuan Ma×1 Yixin Wu×1 Wenbin Liao×1 Feiran Zhang×1 Wenjie Li×1 Lihao Wang×1 Xiaoqing Zheng×1 Xuefeng Yan×1 Lei Bai×1 Ya-Qin Zhang×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.

Name is common (Hao Wang); the profile is name-resolved from public bibliographic data, with no ORCID or institutional affiliation provided.
Radar scores are computed from public bibliometric data only; output, impact, and activity dimensions lack underlying data and are left unscored.
The reproducibility score (20) rests on a single assessed paper; two of the three catalog papers carry no field label, Evolve verdict, keywords, or reproducibility assessment.
The catalog covers only three 2026 arXiv papers; no citation counts, venue record, or career facts are available in this snapshot.