Early-career researcher spanning robot learning for physical AI and trustworthy ML (probability calibration)
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-08) 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 cannot be scored yet: the OpenAlex record holds only 2 works, both dated 2026, with h-index and i10-index recorded at 0, so there is no publication-volume history for this axis to measure.
Impact—
Impact cannot be scored yet: with both cataloged papers dated 2026 and citation-based indicators (h-index 0, i10-index 0) not accrued, there is no citation signal; this axis measures scholarly uptake and influence.
Focus0/100
Focus scores 0 because the record shows no concentration yet: the field distribution splits evenly at 0.5 robotics/embodied AI (robot learning) and 0.5 machine learning (trustworthy AI / probability calibration), and each of the five indexed topics—reinforcement learning in robotics, robot manipulation and learning, AI-based planning, adversarial robustness, and explainable AI—appears only once. With just two cataloged papers in two different fields, no dominant direction has emerged.
Collaboration66/100
Collaboration scores 66 on the strength of 27 distinct coauthors in the recent OpenAlex window and 12 recurring collaborators, including Xin Wang, Wenhao Wu, Menghao Zhang, Zhi Wang, Kun Shao, and Jian Luan. On the two cataloged papers Xie appears in middle authorship on HarnessPAI and last authorship on LWCal, neither as corresponding author, consistent with a contributor role inside larger teams rather than a solo line.
Activity32/100
Activity scores 32, reflecting a thin recent record: only 2 works in the last three years, both dated 2026 in the OpenAlex counts-by-year data. Because the entire indexed output falls in a single year and no earlier years appear in the record, a sustained publication cadence cannot yet be established; this score should be read as an early-stage signal rather than a slowdown.
Reproducibility78/100
Reproducibility is the strongest dimension at 78, computed as the mean over 2 assessed papers. Evolve rates LWCal at high reproducibility readiness—simple method, honest evaluation—while HarnessPAI sits at medium, with strong simulated gains but per-task fitting to test perturbations and thin statistics. A two-paper base is small, but consistent above-average readiness across both assessments supports the score.
Research summary
Yuan Xie is an early-career researcher whose cataloged work spans two distinct areas: robot learning for physical AI, and trustworthy machine learning, specifically probability calibration. Evolve's catalog includes two 2026 arXiv papers. The first, HarnessPAI, proposes an evolving code-level harness around frozen vision-language-action models and world action models, reporting large simulated gains (LIBERO-PRO +61.6), with review caveats about per-task fitting to test perturbations and thin statistics; Xie is a middle author. The second, LWCal, introduces loss-weighted calibration for tabular classifiers with noisy calibration labels, evaluated honestly but only under synthetic noise on small benchmarks; Xie is the last author. Bibliometrically, the OpenAlex record is sparse and recent: 2 works in the last three years, both dated 2026, with h-index and i10-index at 0, so output and impact cannot yet be scored. The field distribution splits evenly (0.5 each) between robotics/embodied AI and machine learning (trustworthy AI), and each of five indexed topics appears once, yielding a focus score of 0. Collaboration is comparatively stronger (66), with 27 distinct coauthors and 12 recurring collaborators. Reproducibility is the highest dimension (78), a mean over two assessed papers whose readiness ranges from high (LWCal) to medium (HarnessPAI). Note that 'Yuan Xie' is a common name; this profile is name-resolved from public bibliographic data, and institution, country, and ORCID are unrecorded.
Collaboration pattern
With 27 distinct coauthors in the recent OpenAlex window and 12 recurring collaborators—Xin Wang, Wenhao Wu, Menghao Zhang, Zhi Wang, Kun Shao, Jian Luan, Yang Li, Qing Li, Shangding Gu, Huichi Zhou, Shuqing Shi, and Fei Ni—Xie works inside sizeable teams. On the two cataloged papers Xie is a middle author on HarnessPAI and last author on LWCal, neither corresponding, matching the collaboration score of 66.
Research keywords
Highlights
Code-evolving harness for physical AI
Loss-weighted calibration under noisy labels
Two-field span: embodied AI and trustworthy ML
The two cataloged papers sit in different fields split 0.5/0.5 in the OpenAlex field distribution—robot learning for physical AI and probability calibration—with each of five topics indexed once, a breadth that drives the focus score to 0.
Timeline
- 2026HarnessPAI: An Evolving Harness for Physical AI posted to arXiv; middle-author position, field: robotics / embodied AI (robot learning).
- 2026LWCal: Loss-Weighted Calibration for Tabular Classifiers with Noisy Calibration Labels posted to arXiv; last-author position, field: trustworthy AI / probability calibration.
- 2026OpenAlex record logs 2 works in the year; both papers enter Evolve's catalog with reproducibility readiness of medium (HarnessPAI) and high (LWCal).
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.
- Bibliometric figures come from OpenAlex only; the record is sparse (2 works, both 2026) with h-index and i10-index at 0.
- Profile is name-resolved from public bibliographic data; 'Yuan Xie' is a common name, and institution, country, and ORCID are not recorded.
- Coverage is limited to the two papers reviewed by Evolve; other publications under this author name may exist but are unassessed.
- The reproducibility score rests on only 2 assessed papers, and the output and impact dimensions are missing for lack of citation data.