Evolve.Pub
Sign inScan your paper
QH

Quan Hao

Beijing University of Technology · China

Unclaimed · AI-structured profileOpenAlex

First-author work spanning AI-driven drug–target interaction prediction and generative augmentation for railway vision

3

papers here

1

reviewed by Evolve

3

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: insufficient dataOutputImpact: insufficient dataImpactFocus: 100/100FocusCollaboration: 58/100CollaborationActivity: 47/100ActivityReproducibility: 55/100Reproducibility

Output—

No output score yet: OpenAlex reports h-index 0 and i10-index 0 with only four indexed works, too little to measure productivity. This axis would track publication volume and venue quality over time.

Impact—

No impact score yet: citation data are insufficient (h-index 0, i10-index 0, arXiv-only 2026 papers). This axis would measure citations, usage, and the influence of the work on subsequent research.

Focus100/100

The focus score of 100 reflects OpenAlex's field distribution assigning full weight to a single field (computational biology / AI-driven drug discovery), with topic counts concentrated on computational drug discovery methods (2), infrastructure maintenance and monitoring (2), and advanced neural network applications (2). The cataloged work unifies these under one methodological agenda: learning-based and generative modeling for scarce-data scientific and industrial perception problems.

Collaboration58/100

Score 58. All three reviewed papers are first-authored with no corresponding-author role recorded; 17 distinct coauthors appear in the recent OpenAlex window and 12 recurring collaborators are listed, including Lechuan Xu, Ziyang Tao, Chenxi Zhang and Yudong Wang. This suggests an active but still-forming network: broad co-author reach relative to a four-paper recent output, typical of an early-career researcher consolidating collaborations within a core group.

Activity47/100

Score 47. OpenAlex counts 4 works in 2026 — the only year with recorded output — matching recent3yWorks = 4, while h-index and i10-index remain 0. All reviewed papers are 2026 arXiv preprints, indicating a recent burst of first-author submissions rather than sustained multi-year volume, consistent with a mid-range activity score for an early-stage profile.

Reproducibility55/100

Score 55, computed as reproMean across 1 repro-assessed paper (reproPapers = 1). ProbeMatchDTI carries medium reproducibility readiness, while RailGen and RailSyn are rated low: both rely on a 102-image validation set, RailGen additionally uses LLM-judge scoring with no released code, and RailSyn omits variance reporting. Improving data release and evaluation rigor would most directly raise this axis.

Research summary

Quan Hao (Beijing University of Technology, China) is an early-stage researcher whose profile is name-resolved from OpenAlex bibliographic data. The catalog shows four works in the last three years, all dated 2026, with first authorship on all three papers Evolve has reviewed. Research splits into two complementary threads. The first is AI-driven drug discovery: ProbeMatchDTI introduces probe-driven multi-scale biochemical pattern matching for drug–target interaction prediction, combining an attention probe mechanism with atom–residue matching. The second is railway infrastructure monitoring: RailGen pairs a multimodal agent with generative data augmentation, including flow matching, to synthesize small foreign objects for intrusion detection, while RailSyn applies diagnosis-guided diffusion-based image generation, reported to lift up to +4.9 AP across nine detectors. Across both threads the unifying theme is generating task-relevant data and representations where real annotations are scarce. OpenAlex metrics depict a focused but very young profile: h-index 0, i10-index 0, a single dominant field assignment, 17 distinct coauthors, and a focus score of 100. Evolve's reviews temper the SOTA claims: near-saturated DTI benchmarks with non-disjoint random splits, 102-image validation sets, LLM-judge scoring, no variance reporting and unreleased code. Reproducibility readiness is currently low-to-medium, and citation-based impact has not yet accrued.

Collaboration pattern

All three cataloged papers are first-authored, with no corresponding-author record, and 17 distinct coauthors appear in the recent OpenAlex window. Twelve recurring collaborators are listed — including Lechuan Xu, Ziyang Tao, Chenxi Zhang, Yudong Wang, Rui Shi and Liguo Zhang — indicating work within a stable core group while reaching a moderately wide network (collaboration score 58). The concentrated 2026 output suggests project-driven team work rather than long-term independent collaborations.

Research keywords

drug–target interaction predictionbiochemical representation learningmulti-scale molecular modelingvirtual screeninggenerative data augmentationdiffusion modelsflow matchingsmall object detectionrailway foreign object detectionvideo surveillancemultimodal agentsattention mechanisms

Highlights

Timeline

  1. 2026Published ProbeMatchDTI on arXiv as first author, applying probe-driven multi-scale biochemical pattern matching to drug–target interaction prediction.
  2. 2026Published RailGen on arXiv as first author, using a multimodal agent to guide generative synthesis of small foreign objects for railway intrusion detection.
  3. 2026Published RailSyn on arXiv as first author, introducing diagnosis-guided diffusion image generation for traceable data completion in railway foreign object detection.
  4. 2026OpenAlex records 4 works in this year — the only indexed publication year so far — with 17 distinct coauthors and h-index 0.

Papers in the Evolve catalog

Frequent co-authors

Lechuan Xu×2Ziyang Tao×2Chenxi Zhang×2Yudong Wang×2Rui Shi×2Liguo Zhang×2Mengyue Fan×1Zifan Dong×1Youru Li×1Jianduo Zhao×1Hao Zhang×1Fei Xia×1

Is this you? Claim this profile

Enter the email you published with. We'll send a one-time claim link; after claiming you can edit the introduction or hide this page. Requests from a mismatched address go to manual review.

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 scores come from OpenAlex only; no ORCID is linked, so the profile is name-resolved from public bibliographic data and may merge or omit same-name records.
  • Output and impact dimensions are unscored because h-index and i10-index are 0 with only four indexed works; citation data have not yet accrued.
  • All cataloged papers are 2026 arXiv preprints; peer-review status and any published versions are unknown.
  • The reproducibility score averages a single assessed paper (reproPapers = 1), so it is coarse.
Quan Hao — Researcher Profile | Evolve.Pub