Peking University LLM-agent researcher co-authoring 2026 work on multi-agent platforms and agentic self-improvement.
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
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 data is insufficient: this OpenAlex record lists no verifiable publication totals, so the axis is unscored. It would measure cumulative works and per-year production once authorship disambiguation is resolved.
Impact—
Impact data is insufficient: recorded h-index and i10-index are both 0 and citation aggregates are absent, likely reflecting sparse OpenAlex attribution rather than measured impact. This axis would measure citation-based influence once the record accumulates.
Focus100/100
A focus score of 100 reflects an extremely concentrated field distribution: the single classified field is AI agents and human-AI collaboration, and all three recorded topics — multi-agent systems and negotiation, AI-based problem solving and planning, and artificial intelligence applications — sit inside the same LLM-agent space, matching the two catalog papers.
Collaboration85/100
The collaboration score of 85 rests on 69 distinct coauthors recorded in OpenAlex — a wide network for a record with so few indexed works. Both 2026 catalog papers are middle-author positions, and the frequent-coauthor list is anchored by an LLM-agent cluster including Yufan Dang, Chen Qian, Zhiyuan Liu, and Tao Gui, consistent with team-based multi-agent research at Peking University.
Activity20/100
Activity is scored 20 on the strength of one work in the last three years and a counts-by-year series whose only entry is 2026 (1 work). Both catalog papers are 2026 arXiv releases, so this reads as an early, front-loaded record with a thin volume signal rather than a declining one.
Reproducibility20/100
The reproducibility score of 20 comes from a reproducibility mean of 20 over one assessed paper. Within the catalog, the Atria Dawn release report is rated low reproducibility readiness and ChatDev 2.0 medium; Evolve's verdicts also flag cherry-picking risk and a lack of statistical rigor in evaluations.
Research summary
Bowen Lai is a researcher at Peking University whose visible record centers on LLM-based multi-agent systems and human-AI collaboration. The Evolve catalog holds two 2026 arXiv papers, both with a middle-author position and no corresponding-author role. The first, 'Atria Dawn: The Dawn of Agentic Superintelligence', is a model release report whose human-AI collaboration study Evolve considers its most interesting component, while flagging cherry-picking risk in its benchmark table and self-reported attribution numbers; it is rated low reproducibility readiness. The second, 'ChatDev 2.0', describes a no-code multi-agent platform with a cycle-aware CADET scheduler that reproduces three workflows at competitive performance, though the evaluation lacks statistical rigor; reproducibility readiness is medium. Bibliometric data is sparse. The OpenAlex record lists 69 distinct coauthors, a single classified field — AI agents and human-AI collaboration — and three topics (multi-agent systems and negotiation; AI-based problem solving and planning; artificial intelligence applications), all within the LLM-agent space. Recorded h-index and i10-index are 0, only one work is counted in the last three years, and the counts-by-year series contains a single 2026 entry, so output and impact axes are unscored. The frequent-coauthor list — including Yufan Dang, Chen Qian, Zhiyuan Liu, and Tao Gui — indicates dense teamwork. This is an early-stage, name-resolved profile based solely on public bibliographic data.
Collaboration pattern
With 69 distinct coauthors recorded against a very short indexed record, Bowen Lai works in large teams. Both catalog papers are middle-author contributions, and the recurring coauthor group — Yufan Dang, Shu Yao, Chenting Xu, Chen Qian, Zhiyuan Liu, Tao Gui, Guanting Dong and others — forms a dense LLM-agent cluster consistent with Peking University's multi-agent systems research.
Research keywords
Highlights
ChatDev 2.0: a no-code multi-agent platform with cycle-aware scheduling
Atria Dawn: agentic self-improvement report with evaluation caveats
Timeline
- 2026Co-authored 'ChatDev 2.0: A No-Code Multi-Agent Platform for Developing Everything' (arXiv), which introduces a cycle-aware CADET scheduler for LLM multi-agent development.
- 2026Co-authored 'Atria Dawn: The Dawn of Agentic Superintelligence' (arXiv), a model release report on an agentic language model centered on recursive self-improvement.
- 2026OpenAlex counts-by-year shows a single indexed work for 2026; the record also lists 69 distinct coauthors, h-index 0, and i10-index 0.
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.
- Bibliometrics are computed from OpenAlex only; the record shows h-index 0, i10-index 0, and a single indexed 2026 work, which may undercount actual output.
- Both catalog papers are arXiv preprints as reviewed by Evolve; the provided data does not reflect peer-reviewed publication status.
- No ORCID is linked; the profile is name-resolved from public bibliographic data, and 'Bowen Lai' is a common name, so misattribution is possible.
- Output and impact dimensions are unscored because citation aggregates are missing from the record.