Kun Kuang

Zhejiang University · China

Unclaimed · AI-structured profileORCID OpenAlex

Causal discovery researcher at Zhejiang University; highly active in 2026 with two reviewed arXiv methods papers

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-09) 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 scorable. The indexed record is sparse — h-index 1 and i10-index 0 under this OpenAlex identity — so no output score is computed. This axis would measure total publication volume, yearly output, and venue spread once sufficient indexed works accumulate.

Impact—

Insufficient data to score. Citation counts have not accumulated for this identity (i10-index 0), so impact is left unestimated. This axis would measure citation-based influence such as total citations, citations per paper, and field-normalized percentiles.

Focus0/100

Scores 0/100 on this scale, which flags extreme concentration rather than absence of work: fieldDist assigns 0.5 to machine learning — causal discovery from time series and 0.5 to machine learning (causal discovery), so recent output collapses into one causal-discovery cluster. Topic counts (topic modeling 7; recommender systems 6; information retrieval 6) suggest adjacent interests, but current flagged work is tightly centered.

Collaboration39/100

Scores 39/100. The snapshot lists 6 distinct coauthors in the recent window, forming a small recurring cluster: Wenbo Xu, Yue He, Yunhai Wang, Yueguo Chen, Zhengkang Guan, and Fei Wu. On both Evolve-reviewed 2026 papers Kuang sits in the last-author position without corresponding-author status, a pattern consistent with senior supervision of a compact team rather than large-scale distributed collaboration.

Activity89/100

Scores 89/100, the strongest axis in the radar. countsByYear rises from 2 works in 2025 to 18 in 2026, and recent3yWorks totals 20, so nearly all indexed output lands in the current year. Both Evolve-reviewed papers — Decoupled Causal Discovery and ITSY — are 2026 arXiv releases inside this surge, marking an intense, ongoing productivity phase.

Reproducibility78/100

Scores 78/100, the mean readiness across 2 Evolve-reviewed papers (reproPapers 2). Decoupled Causal Discovery is rated high readiness while ITSY is medium, so the average rests partly on one strongly reproducible release. Evolve's verdicts also flag evaluation gaps — DCD lacks real-data evidence and ITSY lacks identifiability theory — meaning artifacts look solid but empirical completeness varies across the portfolio.

Research summary

Kun Kuang is a researcher at Zhejiang University whose recent indexed work concentrates on causal discovery in machine learning. The OpenAlex field distribution splits evenly (0.5 / 0.5) between machine learning (causal discovery) and machine learning — causal discovery from time series, and both Evolve-reviewed 2026 arXiv papers — Decoupled Causal Discovery and ITSY — sit squarely in that cluster. Both list Kuang as last author without corresponding status, a placement typical of a senior group lead. Activity is the standout signal: countsByYear rises from 2 works in 2025 to 18 in 2026, recent3yWorks totals 20, and the activity score is 89/100. Focus is extreme (0/100 on the spread scale), while collaboration scores 39/100 across 6 distinct coauthors, including recurring names Wenbo Xu, Yue He, Yunhai Wang, Yueguo Chen, Zhengkang Guan, and Fei Wu. Broader topic counts — topic modeling (7), recommender systems (6), information retrieval and search behavior (6), Bayesian modeling and causal inference (5), multimodal machine learning (5) — indicate adjacent interests around the causal core. Reproducibility averages 78/100 across the 2 reviewed papers. Output and impact axes remain unscored: with h-index 1 and i10-index 0 under this identity, citation-based measures are not yet meaningful.

Collaboration pattern

Kuang's indexed collaboration footprint is compact: 6 distinct coauthors in the recent OpenAlex window and a collaboration score of 39/100. The recurring set — Wenbo Xu, Yue He, Yunhai Wang, Yueguo Chen, Zhengkang Guan, Fei Wu — points to a stable, Zhejiang University-centered group. On both 2026 catalog papers he appears as last author without corresponding status, consistent with a supervisory role in which junior members drive execution while he steers the research direction.

Research keywords

causal discoverystructure learningMarkov boundarysample reweightingstructural causal modelsSVARirregular time seriesmissing data imputationdifferentiable DAG learningcausal discovery from time seriestopic modelingrecommender systems

Highlights

Decoupled Causal Discovery: Markov boundaries via learned sample reweighting
DCD identifies Markov boundaries by learning sample-reweighting that decouples non-target variables, built on a decoupled-distribution formulation with HSIC. Evolve judges the theory sound and the synthetic results strong, but notes the absence of real-data evidence; reproducibility readiness is rated high.
ITSY: causal discovery from irregular time-series data
ITSY handles irregular time series by imputing missing slices through SVAR prediction from the nearest observed history while jointly learning causal graphs with differentiable DAG learning. Evolve finds the approach promising but flags the lack of identifiability theory; reproducibility readiness is medium.

Timeline

20252 works recorded in the OpenAlex countsByYear snapshot — the earliest indexed year in the provided window.
2026Indexed output accelerates sharply: countsByYear reaches 18 works in 2026, and recent3yWorks totals 20.
2026Posts "Decoupled Causal Discovery" to arXiv as last author (non-corresponding); Evolve rates its reproducibility readiness high.
2026Posts "ITSY: Causal Discovery From Irregular Time-Series Data" to arXiv as last author (non-corresponding); Evolve rates its reproducibility readiness medium.

Papers in the Evolve catalog

2026-09-20
Decoupled Causal Discovery
2026-09-20
ITSY: Causal Discovery From Irregular Time-Series Data

Frequent co-authors

Wenbo Xu×1 Yue He×1 Yunhai Wang×1 Yueguo Chen×1 Zhengkang Guan×1 Fei Wu×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.

All bibliometrics come from an OpenAlex identity snapshot; the low h-index (1) and i10-index (0) may reflect incomplete name disambiguation or early indexing rather than measured impact.
Output and impact radar axes are unscored because the underlying citation data is insufficient; no estimates were substituted for them.
Evolve has reviewed only 2 papers from this catalog; topic and field distributions cover the indexed snapshot window only.
Identity is resolved from public bibliographic data under ORCID 0000-0001-7024-9790; no biographical facts beyond the provided snapshot are asserted.