Early-career NLP researcher focused on knowledge-graph and LLM methods for KBQA and scientific hypothesis discovery.
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
1
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 yet scored: the OpenAlex record lacks the complete venue and authorship-role metadata this axis needs. Once filled in, it would measure publication volume, first/corresponding-author share, and publication venues.
Impact—
Not yet scored: citation data are too thin to evaluate — h-index 1 and i10-index 0 give no reliable signal. This axis would measure citation counts, field-normalized influence, and citing works.
Focus100/100
Focus scores 100, the maximum, reflecting an unusually concentrated profile: OpenAlex topic assignments are led by Topic Modeling (14 works), followed by Information Retrieval and Search Behavior (6), Natural Language Processing Techniques (4), Semantic Web and Ontologies (4), and Advanced Graph Neural Networks (4), while the field distribution maps entirely to natural language processing. Both reviewed 2026 papers sit squarely in NLP-knowledge-graph work.
Collaboration51/100
Collaboration scores 51, mid-range: the OpenAlex record lists 12 distinct coauthors, matching Evolve's 12 tracked frequent collaborators, including Yihang Sun, Zhihan Zhu, Zhiyuan Jiang, Jingyi Ge, and Jiaxuan You. On both reviewed 2026 papers Li holds a middle authorship position without corresponding-author status, consistent with contributing within larger, team-led research groups rather than leading projects independently.
Activity84/100
Activity scores 84, high for an early-career record: OpenAlex indexes 17 works over the last three years, with 15 dated 2026 alone, against 2 in 2025 and 1 in 2021. Both catalog entries are 2026 arXiv preprints, consistent with a recent, rapid burst of publishing rather than long-term accumulation.
Reproducibility55/100
Reproducibility scores 55 from a single assessed paper (reproPapers = 1, mean 55). Both reviewed works carry medium repro-readiness: SALR's gains on GrailQA and WebQSP are consistent yet modest with declining IID splits, while HyGRAIL shows promising F1 but inconsistent ablation numbers on one small dataset. The axis needs more assessed papers to stabilize.
Research summary
Zixuan Li is an early-career researcher affiliated with the Chinese Academy of Sciences whose work centers on natural language processing at the intersection of knowledge graphs and large language models. The OpenAlex-based profile shows a highly concentrated research focus (radar score 100): topic modeling is the dominant assigned topic (14 works), followed by information retrieval and search behavior (6), NLP techniques (4), semantic web and ontologies (4), and graph neural networks (4), with the field distribution mapping to natural language processing. The citation record is young — h-index 1, i10-index 0 — but activity is high (radar 84), with 17 works in the past three years and 15 appearing in 2026 alone. Two reviewed 2026 arXiv papers anchor the Evolve catalog: SALR, a schema-anchored latent-reasoning method for semantic-parsing-based knowledge base question answering with consistent but modest controlled gains on GrailQA and WebQSP, and HyGRAIL, a GNN-plus-LLM pipeline for evidence-grounded scientific hypothesis discovery over knowledge graphs with promising F1 but single-dataset evaluation. Collaboration is mid-range (radar 51), with 12 distinct coauthors and recurring collaborators including Yihang Sun, Zhihan Zhu, Zhiyuan Jiang, Jingyi Ge, and Jiaxuan You; Li holds a middle authorship position on both reviewed papers. Output and impact remain unscored pending fuller citation and venue metadata.
Collaboration pattern
Collaboration is mid-range (radar 51): the OpenAlex record lists 12 distinct coauthors, matching Evolve's 12 tracked frequent collaborators, including Yihang Sun, Zhihan Zhu, Zhiyuan Jiang, Jingyi Ge, and Jiaxuan You. On both reviewed 2026 papers Li appears in a middle authorship position without corresponding-author status, suggesting contribution within larger team-led projects rather than solo or senior-led work.
Research keywords
Highlights
SALR: schema-anchored latent reasoning for KBQA
HyGRAIL: GNN-LLM pipeline for KG-based hypothesis discovery
Sharp 2026 output acceleration around a topic-model core
OpenAlex counts 17 works in the last three years, 15 of them in 2026 (2 in 2025, 1 in 2021), with Topic Modeling as the dominant assigned topic (14) — an accelerating, tightly focused early-career record.
Timeline
- 2021First work appears in the OpenAlex record (1 indexed work).
- 20252 works indexed; the record remains sparse before its 2026 expansion.
- 2026Publication surge: 15 works indexed, including two reviewed arXiv papers — SALR for semantic-parsing-based KBQA and HyGRAIL for KG-based hypothesis discovery.
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 metrics (h-index, i10-index, counts by year, topics, coauthor counts) are computed from OpenAlex only.
- Country is listed as US while the affiliation is the Chinese Academy of Sciences; 'Zixuan Li' is a common name, so this profile is name-resolved from public bibliographic data, with ORCID 0009-0007-9962-6917 on file.
- Reproducibility rests on a single assessed paper (reproPapers = 1); output and impact axes are unscored due to insufficient citation and venue metadata.
- Both catalog papers are 2026 arXiv preprints with middle authorship and no corresponding-author role; peer-review status is not reflected in the input data.