AI researcher spanning LLM-based GUI agents and multimodal representation learning via recent 2026 arXiv work
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-19) 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 to score: the OpenAlex snapshot indexes only one work (2026), so cumulative publication volume, venue spread, and yearly trends cannot be computed yet. This axis would measure total research output over time.
Impact—
Impact cannot be scored yet: the snapshot reports h-index 0 and i10-index 0 with a single indexed work, too early to reflect real citation uptake. This axis would measure citations, h-index trajectory, and influence on follow-up research.
Focus0/100
Focus scores 0 because the evidence shows an even split: fieldDist assigns 0.5 to LLM-based GUI agents and 0.5 to computer vision/multimodal machine learning, and all three detected topics — image/video retrieval, GAN-based synthesis, and multimodal applications — carry a count of 1. With only two cataloged papers, no dominant specialization has emerged yet.
Collaboration60/100
Collaboration scores 60, grounded in 19 distinct coauthors in the snapshot and 12 recurring partners, including Bofan Chen, Yongliang Shen, and Wentao Bao, spanning both the GUI-agent and multimodal lines. Both cataloged papers carry a middle author position without corresponding status, so the pattern reads as a contributor inside larger teams rather than an independent group lead.
Activity20/100
Activity scores 20: recent3yWorks is 1, and countsByYear lists a single work in 2026 with no earlier years indexed. Both cataloged papers are 2026 arXiv preprints, so the record is consistent with a recent research start or a lagging index rather than demonstrated sustained multi-year productivity.
Reproducibility55/100
Reproducibility scores 55 as the mean over the 2 reviewed papers (reproMean 55, reproPapers 2), both rated medium readiness. Evolve notes consistent gains and strong ablations for the GUI-agent skill-evolution work but flags missing significance tests, while FLAT's strong ablations coexist with SOTA claims that overreach its own baseline tables.
Research summary
Jun Xiao is an early-stage AI researcher whose profile is name-resolved from public OpenAlex bibliographic data. The cataloged record centers on two 2026 arXiv preprints, in both of which Jun Xiao holds a middle author position without corresponding-author status. The first, Reflect, Revise, Reuse, proposes training-free skill evolution for GUI agents; Evolve's review credits consistent gains and strong ablations while flagging missing significance tests and head-to-head rival baselines. The second, FLAT, trains a single flexible-length continuous embedding that unifies cross-modal retrieval, image captioning, and text-to-image generation; ablations are again strong, though the paper's SOTA claims overreach its own baseline tables. Bibliometrics are thin: the snapshot indexes one work in 2026, reports h-index 0 and i10-index 0, and shows an even 50/50 split between LLM-based GUI agents and computer vision/multimodal machine learning, so no dominant specialization has crystallized. Collaboration is the relative bright spot — 19 distinct coauthors with 12 recurring partners — while recent activity is limited to one work in three years. Reproducibility averages 55 across the two reviewed papers, both rated medium readiness. Because 'Jun Xiao' is a common name and institutional affiliation is unresolved, all claims here are strictly bibliographic.
Collaboration pattern
Collaboration is team-based and broad: 19 distinct coauthors appear in the snapshot, with 12 recurring partners — including Bofan Chen, Boxuan Zhang, Zhengxi Lu, Weiming Lu, Yueting Zhuang, Yongliang Shen, Guangyu Sun, and Shlok Kumar Mishra — spanning both the GUI-agent and multimodal lines. Both cataloged papers are middle-author group efforts rather than solo or corresponding-author work, consistent with a contributor role inside larger teams.
Research keywords
Highlights
Training-Free Skill Evolution for GUI Agents
FLAT: One Flexible-Length Embedding for Retrieval and Generation
Timeline
- 2026Co-authored (middle position) Reflect, Revise, Reuse: Training-Free Skill Evolution for GUI Agents, released on arXiv in the LLM-based GUI agents field.
- 2026Co-authored (middle position) FLAT, an arXiv paper on resampling image and text into 1D flexible-length aligned transmodal tokens for retrieval and generation.
- 2026OpenAlex records this as the only indexed publication year so far (1 work); both papers entered the Evolve review catalog with medium reproducibility readiness.
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.
- All bibliometrics come from the OpenAlex snapshot only; no external citation data was available, and h-index and i10-index both read 0 in this snapshot.
- 'Jun Xiao' is a common name: identity is resolved from public bibliographic data, and institution, country, ORCID, and the Chinese-character name are unconfirmed.
- The output and impact radar axes are marked missing by the radar; they are described qualitatively and deliberately not scored.
- Qualitative assessments cover only the two 2026 arXiv preprints reviewed by Evolve; other works, if any, fall outside this profile's evidence base.