Shaochun Wang

—

Unclaimed · AI-structured profileORCID OpenAlex

Early-stage NLP researcher focused on legal AI for Chinese civil law: GraphRAG reasoning and litigation benchmarks

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—

Not scored: the OpenAlex record is too sparse to establish reliable publication volume beyond scattered per-year counts. This axis would measure total scholarly output, its distribution over years, and authorship roles across venues.

Impact—

Not scored: citation-side data in the record is too thin — an h-index of 1 and an i10-index of 1 are the only signals. This axis would measure scholarly influence through citations, downstream adoption, and uptake of benchmarks and methods.

Focus100/100

Scored 100 — maximal concentration. The field distribution lists exactly one field, natural language processing (legal AI / computational law), and both Evolve-reviewed 2026 papers target Chinese civil law: legal reasoning over the Chinese Civil Code (LEGO) and issue identification in civil litigation (LexIssue). OpenAlex topics reinforce this, with Artificial Intelligence in Law and Topic Modeling leading at 3 hits each.

Collaboration55/100

Scored 55 — moderate, team-based collaboration. The record lists 15 distinct coauthors and 12 recurring collaborators, including Qingjing Chen, Junkai Zhang, Jiahao Ding, Siyuan Zheng, Yukun Yan, and Zhi Zheng. On both Evolve-reviewed 2026 papers Wang appears as a middle-listed author without corresponding-author status, consistent with contributing inside a larger legal-NLP group rather than leading it.

Activity40/100

Scored 40 — modest recent activity. The countsByYear data shows 1 indexed work in 2023, 1 in 2025, and 2 in 2026, with 3 works over the most recent three-year window. The 2026 cluster is the densest and contains both Evolve-reviewed arXiv papers, suggesting an ongoing push in legal NLP rather than a steady multi-year cadence.

Reproducibility55/100

Scored 55, based on a single assessed paper (reproPapers: 1, mean 55). Both Evolve-reviewed 2026 papers carry a medium reproducibility readiness: LEGO's verdict credits measurable same-backbone RAG gains but notes the parity claim with larger models lacks statistical support, while LexIssue's benchmark is flagged for synthetic pleadings and LLM-judge noise. Broader evidence is not yet available.

Research summary

Shaochun Wang (ORCID 0009-0008-0356-8722) is an early-stage researcher whose indexed work concentrates on natural language processing for law, specifically Chinese civil law. The Evolve catalog includes two 2026 arXiv papers with Wang as a middle-listed, non-corresponding author: LEGO, which couples expert-graph GraphRAG with a syllogistic expert chain-of-thought for legal reasoning over the Chinese Civil Code, and LexIssue, a 430-case benchmark for legal issue identification in Chinese civil litigation. Both received 'medium' reproducibility readiness; the LEGO verdict notes strong same-backbone RAG gains while flagging that claimed parity with larger models lacks statistical support, and the LexIssue verdict credits a new benchmark while noting synthetic pleadings and LLM-judge noise cap reliability. The OpenAlex record shows a focus score of 100: the field distribution lists a single field — natural language processing (legal AI / computational law) — with Topic Modeling (3 hits) and Artificial Intelligence in Law (3) as leading topics. Collaboration is moderate (55), with 15 distinct coauthors and 12 recurring collaborators including Qingjing Chen, Junkai Zhang, and Yukun Yan; activity is modest (40), with 3 works in the last three years and per-year counts of 1 (2023), 1 (2025), and 2 (2026). Output and impact are not scored; the h-index and i10-index are both 1, and the record also carries a few physics-oriented topics that may reflect author-disambiguation noise.

Collaboration pattern

Collaboration is team-based and moderately broad: the OpenAlex record lists 15 distinct coauthors, and 12 recur often enough to be flagged, including Qingjing Chen, Junkai Zhang, Jiahao Ding, Siyuan Zheng, Yukun Yan, and Antonino Rotolo. On both Evolve-reviewed 2026 papers Wang is a middle-listed, non-corresponding author, suggesting contribution within a larger legal-NLP group — the shared frequent-coauthor list points to overlapping author pools across the LEGO and LexIssue projects — rather than independent or leading-author work.

Research keywords

GraphRAGlegal reasoningchain-of-thoughtretrieval-augmented generationknowledge graphChinese Civil Codelegal issue identificationChinese civil litigationLLM-as-a-judgebenchmark constructiontopic modelingAI ethics and social impact

Highlights

LEGO: expert-graph GraphRAG plus expert chain-of-thought for legal reasoning
A 2026 arXiv paper pairing expert-graph retrieval-augmented generation with a syllogistic expert chain-of-thought for legal reasoning in Chinese civil law. Evolve's review finds strong same-backbone RAG gains, but judges the claimed parity with larger models to lack statistical support; reproducibility readiness is rated medium.
LexIssue: a 430-case benchmark for legal issue identification in Chinese civil litigation
Introduces a benchmark of 430 cases for identifying legal issues in Chinese civil litigation, evaluated with LLM-as-a-judge and retrieval-augmented pipelines. Evolve's verdict: RAG helps, but synthetic pleadings and judge-model noise cap reliability; reproducibility readiness is medium.
A concentrated legal-AI line takes shape in 2026
Both Evolve-reviewed papers land in 2026 and share one setting — Chinese civil law NLP with retrieval augmentation. This matches the OpenAlex profile, where Artificial Intelligence in Law and Topic Modeling are the top topics with 3 hits each, the field distribution lists a single field, and the focus score is 100.

Timeline

2023Earliest indexed output in the OpenAlex record: countsByYear lists 1 work for 2023.
2025One indexed work recorded for 2025 in the OpenAlex countsByYear data, keeping output sparse between active years.
2026Most productive indexed year: 2 works, including both Evolve-reviewed arXiv papers — LEGO (expert-graph GraphRAG for legal reasoning) and LexIssue (a 430-case litigation benchmark).

Papers in the Evolve catalog

Frequent co-authors

Qingjing Chen×1 Junkai Zhang×1 Jiahao Ding×1 Siyuan Zheng×1 Yukun Yan×1 Zhi Zheng×1 Antonino Rotolo×1 Yun Liu×1 Weixing Shen×1 Huiyuan Xie×1 Yuqin Huang×1 Zhicheng Hao×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.

Bibliometrics come from an OpenAlex record (identity: openalex) linked to ORCID 0009-0008-0356-8722; institution and country are absent from the source data.
This is an early-stage, name-resolved profile built from public bibliographic data; the record's topic list includes physics-oriented topics (superconductivity and magnetism; rare-earth and actinide compounds) outside the legal-AI line, which may indicate homonym blending in author disambiguation.
Output and impact dimensions are unscored: h-index 1 and i10-index 1 are the only citation-side signals in the record.
The reproducibility score rests on a single assessed paper (reproPapers: 1); only two papers have full Evolve reviews.