Fudan University researcher bridging LLM-based social simulation and science-of-science forecasting with LLM agents.
2
papers here
2
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
2
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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 computed: this axis would measure publication volume and productivity over time. OpenAlex currently links only four works in the recent three-year window (2024: 1, 2025: 1, 2026: 2), too thin for a reliable score.
Impact—
Not yet computed: this axis would measure citation uptake and scholarly influence. Available signals are sparse — h-index 2 and i10-index 0 — consistent with an early-career, preprint-heavy record rather than a measurable impact profile.
Focus0/100
The focus score is 0, reflecting an even 50/50 field distribution split between LLM-based computational social science and LLM-agent evaluation / science of science. Topic records (topic modeling x2, scientometrics, multi-agent negotiation, simulation techniques, stock-market forecasting) reinforce this breadth, and the two 2026 catalog papers sit on either side of the divide, so the current agenda reads as dual-track rather than concentrated.
Collaboration66/100
The collaboration score is 66, built on 27 distinct coauthors in the OpenAlex record and 12 recurring collaborators (including Jingcong Liang, Xinnong Zhang, Xinyi Mou, Siyuan Wang, and Hanjia Lyu). Both catalog papers are multi-author efforts — first author on RAP and a middle-position author on SocioVerse2, with no corresponding-author role recorded — a team-based pattern typical of large LLM research groups.
Activity47/100
The activity score is 47. The counts-by-year record shows 1 work in 2024, 1 in 2025, and 2 in 2026, with recent3yWorks at 4 — a thin but accelerating cadence. Both catalog papers appeared on arXiv in 2026, suggesting current effort is concentrated on preprint-stage releases rather than a steady journal or conference stream.
Reproducibility55/100
The reproducibility score is 55, averaged across 2 assessed papers, both rated medium readiness. Evolve's verdicts flag concrete gaps: RAP's experiments rely on model-built instruments with unreleased code, while SocioVerse2's validation partly repackages the authors' companion studies — methods are documented, but independent re-execution is not yet straightforward.
Research summary
Yingqian Wu is an early-career researcher at Fudan University (CN) whose profile, resolved from public OpenAlex bibliographic data, shows a clearly dual-track agenda at the intersection of LLM agents and computational social research. One track builds LLM-based social simulation: the 2026 SocioVerse2 framework introduces longitudinal, dynamic simulation under a human-AI co-evolutionary paradigm, with a seven-case evaluation that Evolve judged solid even while noting its validation leans on the authors' companion studies. The other track applies LLM agents to the science of science: the first-authored RAP (2026) tasks agents with forecasting arXiv field shares and finds they all trail an exponential moving average baseline, framing the gap as target-conditioned evidence acquisition bias. The bibliometric record is thin and recent: four works in the last three years (2024: 1, 2025: 1, 2026: 2), an h-index of 2 with i10-index 0, and both catalog papers still at arXiv preprint stage with medium reproducibility readiness (mean 55 across 2 papers). Collaboration is the strongest measured signal (66), with 27 distinct coauthors and 12 recurring collaborators. Output and impact dimensions are not yet scored, and citation coverage remains sparse — this is a profile whose current value lies in research direction and positioning rather than an accumulated citation footprint.
Collaboration pattern
Collaboration is this profile's strongest measured axis (score 66): 27 distinct coauthors and 12 recurring collaborators, including Jingcong Liang, Xinnong Zhang, Jiayu Lin, Xinyi Mou, Siyuan Wang, and Hanjia Lyu. Both catalog papers are team efforts — first author on RAP, middle-position author on SocioVerse2, with no corresponding-author role recorded — consistent with contributing within a larger LLM research group rather than solo or PI-led output.
Research keywords
Highlights
SocioVerse2: longitudinal social simulation under a human-AI co-evolutionary paradigm
RAP: first-author study shows LLM agents trail a simple baseline in forecasting research attention
A dual-track agenda linking simulated societies to the science of science
The field distribution splits evenly (0.5/0.5) between LLM-based social simulation and science-of-science forecasting, with the two 2026 papers on opposite sides — one simulating society with agents, one using agents to measure science itself.
Timeline
- 2024First work appears in the OpenAlex record — one publication, the earliest trace of this bibliographic identity.
- 2025A single recorded work; output stays sparse ahead of the 2026 preprint push.
- 2026SocioVerse2 posted to arXiv — a longitudinal dynamic social simulation framework under a human-AI co-evolutionary paradigm (middle-author role).
- 2026RAP posted to arXiv as first-author work, showing LLM agents forecasting arXiv field shares all trail an EWMA baseline.
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 indicators come from OpenAlex only; no ORCID is linked, so this profile is name-resolved from public bibliographic data and may merge or split records for a relatively common Chinese name.
- Both catalog papers are 2026 arXiv preprints; peer-reviewed venue acceptance is not reflected in this data.
- Only 2 papers have been reviewed by Evolve, and the output and impact dimensions lack sufficient scoring data.
- Citation coverage is sparse (h-index 2, i10-index 0), so impact-related statements should be read as early signals, not established influence.