Beijing Institute of Technology researcher spanning Shapley-based feature attribution and quantum reservoir computing
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—
No output metrics are available for this profile yet — the snapshot contains no publication counts or venue volume, so the axis is left unscored. It would measure the size and consistency of Yiming Li's research output across years and venues.
Impact—
Impact cannot be scored: no citation counts, h-index, or downstream-uptake evidence exists in the provided snapshot. This axis would measure how often and how widely the community cites and builds on Yiming Li's cataloged papers relative to field norms.
Focus0/100
Focus scores 0 because the field distribution is split exactly in half: 0.5 statistical machine learning (feature attribution and missing data) and 0.5 quantum machine learning. The two cataloged 2026 papers sit in disjoint problem spaces — Shapley estimation for multi-site blockwise-missing data versus Floquet-engineered quantum reservoir computing for molecular properties — with no shared methodology visible, indicating broad rather than concentrated focus.
Collaboration50/100
Collaboration scores 50 on 11 distinct coauthors in the raw record, with frequent partners including Siqi Li, Wangxuan Fan, Doudou Zhou, Molei Liu, and Zhang-Qi Yin. On both cataloged papers Yiming Li is a middle author and not corresponding, indicating contribution within larger teams rather than solo-led projects; the mid-range score reflects a genuine but still limited visible collaboration network.
Activity—
Activity is unscored because the record holds only two 2026 arXiv papers with no multi-year publication cadence. It would measure recent publishing tempo, sustained engagement over time, and whether output is accelerating or pausing.
Reproducibility100/100
Reproducibility scores 100, the mean across the 2 papers Evolve has reviewed (reproMean 100 over reproPapers 2), both rated high readiness. The FUSHAP paper reports strong results via influence-function control variates despite informal theory, while the quantum reservoir work offers rigorous comparisons against matched echo state networks plus a small hardware demo. A perfect mean on a two-paper sample is promising but preliminary.
Research summary
Yiming Li is a machine learning researcher affiliated with Beijing Institute of Technology whose visible work splits evenly between two lines of inquiry. The first, in statistical machine learning, addresses feature attribution and missing data: the 2026 arXiv paper 'Shapley Value Estimation for Multi-Site Data with Blockwise-Missing Features' (FUSHAP) estimates Shapley values without imputation across multi-site datasets using influence-function control variates, delivering strong empirical results although the supporting theory remains informal. The second, in quantum machine learning, applies Floquet-engineered reservoirs to chemistry: 'Coherent Floquet quantum reservoirs for molecular property prediction' shows discrete-time-crystal quantum reservoir computing with controlled resets beating matched echo state networks on molecular classification and gap forecasting, plus a small hardware demonstration, though graph-neural-network baselines are absent. Evolve has reviewed both papers and rates each high in reproducibility readiness, yielding a reproducibility score of 100 over the two-paper sample. On the cataloged record Li appears as a middle author, never corresponding author, working within a network of 11 distinct coauthors. This profile is name-resolved from public bibliographic data; output, impact, and activity metrics are not yet available and are left unscored.
Collaboration pattern
Yiming Li's cataloged work is team-based: 11 distinct coauthors appear in the raw record, and on both 2026 papers Li is a middle author rather than first or corresponding. The frequent-coauthor list spans both research lines, including Siqi Li, Wangxuan Fan, Doudou Zhou, Molei Liu, and Zhang-Qi Yin, though the input data does not attribute specific coauthors to specific papers.
Research keywords
Highlights
Imputation-free Shapley estimation for multi-site blockwise-missing data
Coherent Floquet quantum reservoirs beat matched echo state networks
Timeline
- 2026Publishes 'Shapley Value Estimation for Multi-Site Data with Blockwise-Missing Features' (FUSHAP) on arXiv, applying influence-function control variates to feature attribution under multi-site blockwise missingness.
- 2026Publishes 'Coherent Floquet quantum reservoirs for molecular property prediction' on arXiv, showing discrete-time-crystal quantum reservoir computing with controlled resets outperforming matched echo state networks.
- 2026Both cataloged papers enter the Evolve review with high reproducibility readiness, yielding a reproducibility score of 100 across the two-paper sample.
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 bibliometric signals derive from the Evolve-reviewed snapshot (2 arXiv papers); no external citation database record was provided.
- Output, impact, and activity dimensions are unscored because publication counts, citation metrics, and multi-year cadence are absent from the input data.
- 'Yiming Li' is a common name; this profile is name-resolved from public bibliographic data and the Chinese character form of the name is unconfirmed.
- The reproducibility score of 100 is based on only 2 reviewed papers and should be read as preliminary.