Huawei AI researcher working across LLM agent self-evolution and large language models for medical time series
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
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—
Output volume is not yet computable: this profile rests on only two reviewed arXiv papers, with no total publication count or venue history. This axis would measure yearly paper output and its trajectory once a fuller publication record is linked.
Impact—
Impact cannot be scored: no citation counts, h-index, or downstream-usage signals are present in the input data. This axis would measure citations, adoption of TTSE's dual-track memory design, and uptake of the medical time-series taxonomy once citation data is linked.
Focus0/100
The focus score is 0/100, reflecting an even field split: the field distribution assigns 0.5 to machine learning (LLM agent self-evolution) and 0.5 to medical AI (biomedical signal processing × NLP). With only two reviewed papers, one in each area, this reads as an intentionally broad, two-track agenda rather than a settled specialty; more papers would clarify whether one line becomes dominant.
Collaboration56/100
Collaboration scores 56/100, with 16 distinct coauthors across the reviewed work. A recurring pool includes Ruimin Pei, Yongkang Wu, Shangyi Zheng, Yaqing Zhang, Deyang Li, Jianjun Tao, Xinyu Zhang, Yu Han, Cigdem Beyan, Xiaofeng Liu and Nan Liu. Authorship positions — last on TTSE, middle on the medical review, never corresponding — fit a senior-contributor role inside team-based industry research rather than solo or small-group work.
Activity—
Activity is not yet measurable: no publication-date series beyond two 2026 papers and no conference talk or dataset-release signals exist in the data. This axis would measure publishing cadence and engagement over time once a longer record is available.
Reproducibility55/100
Reproducibility scores 55/100 as the mean over the 2 papers with reproducibility assessments, both rated medium readiness. For TTSE, Evolve notes a clean risk theory but flags that a pivotal ablation lacks significance and the experience warehouse drives most of the SOPBench gain; the medical review is descriptive over 62 studies. Reproducibility evidence is real but limited to this small sample.
Research summary
Xiang Zhang is an AI researcher at Huawei Technologies (China) whose reviewed work to date spans two distinct lines: machine learning on LLM agent self-evolution and medical AI combining biomedical signal processing with NLP. The Evolve catalog holds two 2026 arXiv papers. On the agent side, TTSE: A Two-Track Online Self-Evolution Framework — where Zhang is the last author — proposes FACT/TIP dual-track memory separating declarative from procedural knowledge, with rule lifecycle management and a clean risk theory for online agent evolution, though a pivotal ablation lacks significance and the experience warehouse drives most of the SOPBench gain. On the medical side, Zhang is a middle author on a clinically oriented review of large language models for medical time series (ECG, EEG, PPG, wearable sensors), offering a taxonomy and guidance built on 62 studies. Field distribution splits evenly (0.5/0.5) between the two areas, and the radar places focus low while collaboration (56/100, 16 distinct coauthors) and reproducibility (55/100 over 2 papers, both medium readiness) sit mid-range. This is a full but early-visibility profile built on a narrow evidence base.
Collaboration pattern
Zhang's recorded network spans 16 distinct coauthors, with a stable recurring pool: Ruimin Pei, Yongkang Wu, Shangyi Zheng, Yaqing Zhang, Deyang Li, Jianjun Tao, Xinyu Zhang, Yu Han, Cigdem Beyan, Xiaofeng Liu and Nan Liu. Authorship positions (last on TTSE, middle on the medical review, neither corresponding) suggest a senior contributor inside team-based industry research at Huawei, active across both the agent and medical-AI lines.
Research keywords
Highlights
TTSE: dual-track online self-evolution for LLM agents
Review maps LLMs onto medical time series
Two-track research range: agents and clinical signals
The field distribution splits 0.5/0.5 between LLM agent self-evolution and medical AI, and the 16-coauthor network spans both lines — a breadth rare at this early-visibility stage of a profile.
Timeline
- 2026TTSE: A Two-Track Online Self-Evolution Framework appears on arXiv, with Zhang in the last-author position.
- 2026Large language models in medical time series analysis appears on arXiv, with Zhang as a middle author.
- 2026The same year's output spans two fields at a 0.5/0.5 split, drawing on 16 distinct coauthors, with a reproducibility mean of 55 over the 2 assessed papers.
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.
- Name-resolved from public bibliographic data: 'Xiang Zhang' is a common name and no ORCID is provided, so attribution to this specific individual carries residual ambiguity.
- Bibliometric scores and the reproducibility mean are computed from a very small base: two reviewed 2026 arXiv papers only.
- Output, impact and activity axes are missing — no citation counts, publication totals, or multi-year record are available in the input data.
- Both papers are arXiv preprints (2026); peer-review status, venue acceptance, and career history beyond 'Huawei Technologies (China)' are not captured.