Bo Li

University of Macau

Unclaimed · AI-structured profileOpenAlex

ML researcher spanning AI agents, spatio-temporal traffic forecasting, and computational biology

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

—

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-02) 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: this profile currently rests on three 2026 arXiv preprints, with no publication counts or venue history available. This axis would measure cumulative output volume and its distribution across years and publication venues.

Impact—

Not yet scored: no citation counts, h-index, or citation-trajectory data exist for this record. This axis would measure the citation-based influence of the researcher's work within and across fields over time.

Focus0/100

Sits at 0/100 on the focus scale: the field distribution splits exactly one third each across AI agents and human-AI collaboration, spatio-temporal machine learning for intelligent transportation, and computational biology for perturbation prediction. With one paper per field among the three catalog entries, this reads as a deliberately broad, exploration-stage portfolio rather than a single-track agenda; convergence may emerge as more work is cataloged.

Collaboration86/100

Scores 86/100 on strong collaboration evidence: 72 distinct coauthors appear in the record, with 12 recurring collaborators including Honglin Guo, Tao Gui, Yicheng Chen, and Guanting Dong. All three catalog papers carry middle-position, non-corresponding authorship, consistent with a well-connected contributor working inside larger teams rather than leading small author groups; the broad coauthor pool mirrors the three-field spread.

Activity—

Not yet scored: annual activity cannot be computed from a snapshot containing only 2026 preprints. This axis would measure recent publication cadence and sustained research engagement over successive years.

Reproducibility43/100

Scores 43/100, computed as the mean repro-readiness across the three papers assessed by Evolve (reproPapers: 3). Readiness is medium for the LoReST traffic-forecasting study and the CELLAUDIT input-use audit, and low for the Atria Dawn model report, whose self-reported benchmark numbers carry cherry-picking risk. The pattern suggests methods-heavy papers with moderate artifact support, but none yet rated highly reproducible.

Research summary

Bo Li (name resolved from public bibliographic data; 'Bo Li' is a common name and no ORCID or affiliation is on record) appears in the Evolve catalog through three 2026 arXiv preprints spanning three fields in equal measure: AI agents and human-AI collaboration, spatio-temporal machine learning for intelligent transportation systems, and computational biology for perturbation prediction. On all three papers Bo Li is a middle-position author without corresponding status, consistent with a contributing member of larger teams. The reviewed work shows a recurring concern with evaluation rigor. LoReST claims state-of-the-art large-scale traffic forecasting on the LargeST benchmark via relation-aware local aggregation and cross-region attention, but on single-seed evidence; CELLAUDIT audits whether agent-discovered cell models genuinely use perturbation inputs and is judged rigorous and honest, though its measured effects are tiny and its sci-Plex gains underpowered; the Atria Dawn release report contains a notably interesting human-AI collaboration study alongside benchmark tables flagged for cherry-picking risk. Quantitatively, the collaboration dimension scores 86/100 on 72 distinct coauthors with 12 recurring collaborators; reproducibility averages 43/100 across the three assessed papers (readiness medium, medium, low); focus sits at 0/100, reflecting an even three-way field split. Output, impact, and activity are not yet scored for lack of underlying data.

Collaboration pattern

Bo Li's collaboration footprint is broad: 72 distinct coauthors, with 12 recurring collaborators including Honglin Guo, Tao Gui, Yicheng Chen, Guanting Dong, Qiming Ge, Yuyang Hu, Zixian Huang, Jiajie Jin, Alexander Lam, Yining Li, Jiahang Lin, and Yanjiang Liu. Across all three catalog papers Bo Li sits in a middle author position without corresponding status, suggesting a dependable contributor inside larger project teams rather than a lone-lead pattern; the collaboration dimension scores 86/100.

Research keywords

traffic forecastingspatio-temporal graph neural networkscross-region attentionspatial heterogeneityAI virtual cellsperturbation response predictioninput-use auditingagentic model discoveryLLM agentshuman-AI collaborationtool-using agentsbenchmark evaluation

Highlights

LoReST: Claiming SOTA Large-Scale Traffic Forecasting
The 2026 preprint reports state-of-the-art large-scale traffic forecasting on the LargeST benchmark, using relation-aware local aggregation and cross-region attention to capture spatial heterogeneity; Evolve's review finds the results strong but flags that evidence rests on single-seed runs, with medium reproducibility readiness.
CELLAUDIT: Falsifying Input-Use Claims in Agent-Discovered Cell Models
This audit framework traces claims from source code to predictive contribution, testing whether AI virtual cell models truly use perturbation inputs; Evolve judges it rigorous and honest, while noting the measured effects are tiny and the sci-Plex gains underpowered, with medium reproducibility readiness.
Atria Dawn: An Agentic LM Report with a Standout Collaboration Study
The release report's most interesting part is its human-AI collaboration study, but Evolve flags cherry-picking risk in the benchmark table and self-reported attribution numbers; it is the least reproducible of the three catalog papers, rated low.

Timeline

2026Co-authored (middle author) the LoReST preprint on large-scale traffic forecasting, built on relation-aware local aggregation and cross-region attention.
2026Co-authored CELLAUDIT, auditing input-use claims in agent-discovered cell models from source code to predictive contribution.
2026Co-authored the Atria Dawn agentic language model release report, which includes its human-AI collaboration study.

Papers in the Evolve catalog

Frequent co-authors

Honglin Guo×1 Tao Gui×1 Yicheng Chen×1 Guanting Dong×1 Qiming Ge×1 Yuyang Hu×1 Zixian Huang×1 Jiajie Jin×1 Alexander Lam×1 Yining Li×1 Jiahang Lin×1 Yanjiang Liu×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.

Bibliometric dimensions are computed from OpenAlex-derived public bibliometric data only; no citation, venue, or h-index metrics were available.
'Bo Li' is a common name; this profile is name-resolved from public bibliographic data and lacks an ORCID, so it may merge or split same-name researchers.
The reviewed catalog covers only three 2026 arXiv preprints; output, impact, and activity dimensions lack underlying data and are unscored.
Institution, country, and career details are not on record, so no affiliation-based context is provided.