Oxford researcher bridging AI and computational immunology: discovery foundation models and immune world modeling
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-15) 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 snapshot contains no aggregate works count, only two 2026 arXiv preprints, and h-index and i10-index both stand at 0 — an early OpenAlex footprint rather than a measured productivity level. This axis would measure total publication volume and structural productivity.
Impact—
Impact cannot be scored: no citation counts, h-index signal, or venue acceptance data exist for these two 2026 preprints. This axis would measure citation uptake, influence on follow-up work, and community recognition once the papers begin to accrue citations.
Focus0/100
The focus score is 0, and the raw data explain why: the field distribution splits exactly 50/50 between artificial intelligence and computational biology / machine learning for immunology, and all five indexed topics carry a count of 1 — no dominant theme has emerged. The strands are connected (AI methods applied to immune modeling), but measured concentration is genuinely low at this stage.
Collaboration54/100
The collaboration score of 54 rests on 14 distinct coauthors in the raw snapshot. Recurring names include Ling Yang and Yingcheng Wu, alongside Taoyong Cui, Xi Wang, Zonghang Li, Jinchao Ding, Fang Wu, and Kejun Ying. Both cataloged papers carry a middle authorship position with no corresponding-author role, indicating contribution within larger team-led groups rather than an independent lead.
Activity32/100
The activity score of 32 reflects a very short visible record: countsByYear shows 2 works in 2026 and nothing earlier, and recent3yWorks is 2. All indexed output appeared in a single year, so there is no multi-year cadence to evaluate; the score will firm up as later years accumulate.
Reproducibility20/100
The reproducibility score of 20 comes from a reproMean of 20 over 2 assessed papers. Both cataloged works rate low on reproducibility readiness: the Discovery Foundation Models paper implements and measures nothing, and the immune world model paper reports disciplined gating but unreleased code. No artifacts accompanied either preprint at review time.
Research summary
Zhenfei Yin is a University of Oxford–affiliated researcher whose 2026 output sits at the intersection of artificial intelligence and computational immunology. The available bibliometric snapshot, resolved through OpenAlex, records two arXiv preprints in 2026, with the field distribution split evenly — 50% artificial intelligence and 50% computational biology / machine learning for immunology — and five topics each indexed once, from cancer immunotherapy biomarkers to AI-based problem solving and logic and reasoning. The first cataloged paper, 'Discovery Foundation Models: Toward Open-Ended Discovery Intelligence', sets out a taxonomy and process protocol for open-ended AI-driven scientific discovery; Evolve's review finds the framework ambitious and the taxonomy useful, but notes that nothing is implemented, measured, or shown to transfer. The second, 'An immune world model for multiscale forecasting and therapeutic hypothesis generation', uses an AI-Scientist-built world model to forecast multiscale immune states and nominate IL-36γ+SIRPα as a therapeutic hypothesis, with disciplined gating but unreleased code and modest reported gains. Yin appears in a middle authorship position on both, never as corresponding author, within a network of 14 distinct coauthors including recurring collaborators such as Ling Yang and Yingcheng Wu. Quantitative impact indicators are not yet available — h-index and i10-index both stand at 0 in the snapshot — so the output and impact axes are marked missing rather than scored. This is an early-stage profile: a coherent direction and recent, team-embedded output, but limited independent evidence so far.
Collaboration pattern
Zhenfei Yin's visible work is team-embedded: the raw snapshot lists 14 distinct coauthors, with Ling Yang and Yingcheng Wu recurring most often, alongside Taoyong Cui, Xi Wang, Zonghang Li, Jinchao Ding, Lingsen You, Yuzhi Xu, Wanghan Xu, Fang Wu, and Kejun Ying. On both cataloged papers Yin holds a middle, non-corresponding authorship position, consistent with a contributing specialist within larger AI-for-science groups rather than an independent lab lead.
Research keywords
Highlights
Discovery Foundation Models: a framework without an implementation
An AI-Scientist-built immune world model
Timeline
- 2026Co-authored 'Discovery Foundation Models: Toward Open-Ended Discovery Intelligence' (arXiv), a framework paper on open-ended discovery intelligence.
- 2026Co-authored 'An immune world model for multiscale forecasting and therapeutic hypothesis generation' (arXiv), applying an AI-Scientist-built model to multiscale immune state forecasting.
- 2026Both works enter the Evolve catalog; with 2 works in countsByYear, recent3yWorks of 2, and an h-index of 0, the profile stands at an early stage.
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 are drawn from OpenAlex only; identity is resolved via the OpenAlex author entity with no ORCID supplied, so name-collision risk cannot be fully excluded.
- h-index and i10-index are 0 in the snapshot and no citation counts exist for the two 2026 preprints; the output and impact axes are therefore marked missing, not scored.
- Coverage is limited to the two Evolve-reviewed 2026 arXiv preprints; no peer-reviewed publications, grants, awards, or career history are in scope.
- Both papers are middle-authorship preprints rated low reproducibility readiness (code unreleased / no implementation), so per-paper claims reflect the reviews, not independent verification.