Xiaochun Yang

Northeastern University · United States

Unclaimed · AI-structured profileORCID OpenAlex

Knowledge-grounded NLP researcher; last author on 2026 event linking and LLM-based knowledge graph completion papers

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-17) 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 — cumulative publication volume and year-by-year productivity — is not yet scoreable for this profile: the OpenAlex identity records only recent activity (18 works in 2026), with no longer-run publication trajectory available to aggregate.

Impact—

Impact — citation accumulation, h-index, and community uptake — is not scoreable yet: the only signals are an h-index of 1 and an i10-index of 0, too thin to characterize influence, so no estimate is offered.

Focus100/100

Focus scores 100/100. The OpenAlex field distribution places this identity wholly in natural language processing, and both 2026 catalog papers target knowledge-grounded NLP tasks — event linking and LLM-based knowledge graph completion. Topic counts concentrate on topic modeling (10) and advanced graph neural networks (9), with semantic web and ontologies (6), recommender systems (5), and information retrieval (4) as adjacent areas.

Collaboration46/100

Collaboration scores 46/100, with 9 distinct coauthors recorded for this identity. The frequent-coauthor list — Yinan Liu, Ziyang Zhang, Boyi Xue, Yingxuan Huang, Bin Wang, Jiaqi Wang, Dongying Lin, and Yang Yang — is modest in scale. Both catalog papers place this researcher in the last author position (non-corresponding), a senior-advisor pattern on small teams rather than a large distributed network.

Activity86/100

Activity scores 86/100. OpenAlex countsByYear records 18 works in 2026 — the only year present — and this equals the recent3yWorks figure of 18, meaning all recorded output falls inside the recent window. Combined with two arXiv papers in the 2026 Evolve catalog, this points to a highly active, current production phase.

Reproducibility55/100

Reproducibility scores 55/100, computed from one assessed paper (reproMean 55, reproPapers 1). That paper — the event linking study — carries Evolve's medium reproducibility readiness; its verdict notes gains that are credible but partly due to floor effects and a high online cost, which affects the ease of independent replication.

Research summary

Xiaochun Yang (Northeastern University, US; ORCID 0000-0002-6184-4771) is a natural language processing researcher whose recent work centers on knowledge-grounded NLP: linking events to knowledge bases and completing knowledge graphs with large language models. This is an early-stage profile name-resolved from public OpenAlex bibliographic data and two 2026 arXiv papers in the Evolve catalog, on both of which Yang holds the last (senior, non-corresponding) author position. The first paper introduces MACE, an LLM multi-agent evidence-refinement plug-in that lifts three event linkers on two benchmarks; Evolve judges the gains credible but partly due to floor effects, with high online cost, and rates reproducibility readiness as medium. The second combines discrete structural coding with similar-entity information for LLM-based knowledge graph completion and has not yet been reviewed by Evolve. Bibliometric signals show a strongly current record: OpenAlex counts 18 works in 2026 — the only year present and equal to the recent three-year count — driving an activity score of 86/100, while the field distribution places the identity wholly in natural language processing, giving a focus score of 100/100. Topic counts concentrate on topic modeling (10) and advanced graph neural networks (9), with semantic web and ontologies (6), recommender systems (5), and information retrieval (4) adjacent. Collaboration is moderate (46/100) across 9 distinct coauthors; citation-based impact is not yet scoreable (h-index 1, i10-index 0).

Collaboration pattern

The OpenAlex record lists 9 distinct coauthors for this identity. Frequent collaborators include Yinan Liu, Ziyang Zhang, Boyi Xue, Yingxuan Huang, Bin Wang, Jiaqi Wang, Dongying Lin, and Yang Yang. In both 2026 catalog papers Yang holds the last author position (non-corresponding), consistent with a senior supervisory role on small NLP teams; the collaboration score of 46/100 reflects a moderate, concentrated co-author network rather than a large distributed one.

Research keywords

Event linkingCandidate event acquisitionMulti-agent LLM systemsKnowledge base groundingEvent argument extractionKnowledge graph completionTopic modelingGraph neural networksSemantic web and ontologiesRecommender systemsInformation retrievalRetrieve-and-rank

Highlights

MACE plug-in lifts three event linkers across two benchmarks
As last author on this 2026 arXiv study, Yang's team proposes MACE, an LLM multi-agent evidence-refinement plug-in for event candidate acquisition. Evolve's review finds the gains credible but partly attributable to floor effects, with high online inference cost; reproducibility readiness is rated medium.
Discrete structural coding plus similar-entity information for LLM-based knowledge graph completion
A second 2026 arXiv paper, again with Yang as last author, combines discrete structural coding with similar entity information for LLM-based knowledge graph completion. It is in the Evolve catalog but not yet reviewed — no verdict, keywords, or reproducibility assessment are available.
A single assessed paper anchors the reproducibility signal
Evolve's reproducibility dimension for this profile rests on one assessed paper (reproMean 55, reproPapers 1): the event linking study, rated medium readiness — reflecting a plug-in design that is conceptually portable but costly to run online.

Timeline

2026arXiv preprint 'Enhancing Event Candidate Acquisition for Event Linking' appears, with Yang in the last (senior, non-corresponding) author position.
2026Evolve reviews the event linking paper: the MACE plug-in lifts three event linkers on two benchmarks; gains credible but partly floor effects, online cost high.
2026A second arXiv paper on LLM-based knowledge graph completion, combining discrete structural coding with similar entity information, again lists Yang as last author.
2026OpenAlex countsByYear records 18 works in 2026 — the only year with recorded output, equal to the recent three-year count.

Papers in the Evolve catalog

2026-09-12
Enhancing Event Candidate Acquisition for Event Linking

Frequent co-authors

Yinan Liu×2 Ziyang Zhang×1 Boyi Xue×1 Yingxuan Huang×1 Bin Wang×1 Jiaqi Wang×1 Dongying Lin×1 Yang Yang×1 Bin Wang×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 derive from a single OpenAlex identity; citation indices are sparse (h-index 1, i10-index 0), which may reflect early-stage output or incomplete author disambiguation.
'Xiaochun Yang' is a common name; this profile is name-resolved from public bibliographic data and should be verified against ORCID 0000-0002-6184-4771.
Only one catalog paper carries an Evolve verdict and reproducibility assessment; the knowledge graph completion paper is unreviewed, with no verdict, keywords, or slug.
Output and impact dimensions are unscored due to insufficient data, and year-by-year coverage is limited to 2026, so no long-run trajectory can be described.