AI researcher focused on multi-agent vision-language systems and first author of a conceptual shared-memory framework
1
papers here
1
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-20) 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—
Publication-volume data is not available in the current snapshot, so output cannot be scored. This axis would measure the number and cadence of peer-reviewed papers and preprints across the researcher's career.
Impact—
No citation counts, h-index, or venue-level influence metrics are available yet, so impact is unscored. This axis would measure how widely the researcher's work is cited and adopted by other teams.
Focus100/100
Focus scores 100/100 because the field distribution in the snapshot contains exactly one field: 'artificial intelligence — multi-agent systems and vision-language models' (1 paper, 100% of cataloged output). All reviewed work concentrates on shared memory, memory hierarchies, agent collaboration, and provenance for multi-agent VLM teams, indicating a tightly concentrated — though so far very narrow — topical identity.
Collaboration36/100
The bibliometric snapshot records 5 distinct coauthors, all clustered around the single cataloged 2026 paper on collaborative memory for multi-agent VLM systems, yielding a collaboration score of 36/100. The team — Shao-Jun Xia, Di Wang, Liangxi Liu, Hainan Xiong, and Zihao Wang — is real but compact, and a one-paper footprint limits what can be inferred about sustained collaborative networks.
Activity—
Longitudinal publication-frequency data is missing, so recent research activity cannot be scored. This axis would measure output cadence and engagement over the most recent years of the researcher's record.
Reproducibility20/100
Reproducibility scores 20/100, computed from a reproducibility mean of 20 across 1 reviewed paper. The sole cataloged work, 'Collaborative Memory for Multi-Agent VLM Systems', is rated low for repro readiness: Evolve's verdict describes a well-argued but purely conceptual framework with no implementation or empirical validation, so there is currently no experimental basis on which others could replicate the work.
Research summary
Huixin Zhang is an early-stage artificial-intelligence researcher working at the intersection of multi-agent systems and vision-language models (VLMs). The Evolve catalog contains one reviewed paper: 'Collaborative Memory for Multi-Agent VLM Systems' (arXiv, 2026), on which Zhang is first author but not corresponding author. The paper proposes a design framework in which multi-agent VLM teams share a collaborative visual memory, organized through a memory hierarchy with provenance tracking to support agent collaboration. Evolve's verdict characterizes it as well-argued but purely conceptual, with no implementation or empirical validation; accordingly, its reproducibility readiness is rated low, and the reproducibility dimension scores 20/100 (repro mean 20 across 1 paper). The profile is extremely focused: the field distribution contains a single field — artificial intelligence, multi-agent systems and vision-language models — yielding a focus score of 100/100. Collaboration is real but compact, with 5 distinct coauthors (Shao-Jun Xia, Di Wang, Liangxi Liu, Hainan Xiong, Zihao Wang) and a collaboration score of 36/100. Output, impact, and activity cannot be scored because the underlying bibliometric data is not yet available. No institution, country, or ORCID is attached to this record, so the profile is name-resolved from public bibliographic data and should be read as an early-stage profile likely to change as more outputs appear.
Collaboration pattern
All 5 distinct coauthors in the record — Shao-Jun Xia, Di Wang, Liangxi Liu, Hainan Xiong, and Zihao Wang — appear around the single 2026 multi-agent VLM paper, where Huixin Zhang is first author (not corresponding). The collaboration score of 36/100 reflects a real but compact team on a one-paper footprint; no recurring multi-project partnerships can yet be established from the available data.
Research keywords
Highlights
Shared visual memory as a design framework for multi-agent VLM teams
A conceptual contribution not yet backed by implementation
A tightly focused early profile: one field, one paper, five coauthors
Timeline
- 2026First-author preprint 'Collaborative Memory for Multi-Agent VLM Systems' appears on arXiv, proposing shared memory for multi-agent VLM systems.
- 2026Evolve reviews the paper, rating its reproducibility readiness low and noting the absence of implementation and empirical validation.
- 2026Works with five coauthors — Shao-Jun Xia, Di Wang, Liangxi Liu, Hainan Xiong, and Zihao Wang — on the cataloged multi-agent VLM paper.
Papers in the Evolve catalog
- first author
Collaborative Memory for Multi-Agent VLM Systems
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 scores come from a narrow snapshot; the output, impact, and activity dimensions lack underlying data and are not scored.
- Only one paper (a 2026 arXiv preprint) is in the reviewed catalog; the profile is name-resolved from public bibliographic data and should be treated as early-stage.
- No institution, country, or ORCID is available, so identity disambiguation relies solely on bibliographic records.
- Field distribution and reproducibility are computed from a single paper, so both scores are highly sensitive to future additions.