Early-stage researcher in fairness-aware model compression and agentic LLM automation for efficient AI hardware
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—
Data insufficient: only one reviewed 2026 preprint is cataloged, so no output metric can be computed. This axis would measure publication volume, venue tier, and authorship roles once a fuller record is linked.
Impact—
No citation or impact data is available in the snapshot, and none is estimated here. This axis would measure citations, aggregate citation-style indicators, and downstream uptake of the work once citation records accrue.
Focus100/100
Focus scores 100, the maximum, because the field distribution contains exactly one domain: machine learning scoped to fairness-aware model compression and efficient AI hardware. Every cataloged work — currently a single 2026 first-author preprint — falls inside it, so topical concentration is total. With only one paper on record, perfect concentration is partly an artifact of sample size and should be re-read as the portfolio diversifies or confirms this direction.
Collaboration14/100
Collaboration scores 14, derived from exactly one distinct coauthor in the record: Yiyu Shi, the sole recurring collaborator and coauthor of the reviewed 2026 preprint. The pattern is a compact two-person collaboration with Guo in the first-author seat (non-corresponding). It is best read as an early-stage signal rather than an established network profile; the score should rise or diversify as more outputs are indexed.
Activity—
Insufficient data: the record spans a single 2026 preprint, so recent publishing cadence and continuity cannot be characterized. This axis would measure output regularity over recent years once more dated records exist.
Reproducibility55/100
Reproducibility scores 55, computed as the mean over one assessed paper — FairCompressAgent — which Evolve rates medium in reproducibility readiness. The review credits credible search-efficiency gains but flags that FPGA deployment claims are untested and that comparisons rest on three runs, leaving empirical robustness partially open. With a single-paper basis, this mid-range score is a starting point rather than a stable estimate.
Research summary
Yuanbo Guo (ORCID 0000-0002-1309-7572) is an early-stage researcher whose public record in the Evolve catalog currently centers on one first-author paper: the 2026 arXiv preprint FairCompressAgent, an agentic framework for fairness-aware model compression targeting FPGA deployment. The work frames compression as a search problem: an LLM-based agent selects among neural network pruning, quantization, and sparse low-rank factorization under an equalized-opportunity fairness constraint, aiming to produce compressed models suited to efficient AI hardware without disproportionately degrading outcomes for protected groups. Evolve's review judges the framework's search-efficiency gains credible while noting that the FPGA claims remain untested and that comparisons rest on three runs, yielding a medium reproducibility-readiness rating (reproMean 55 over one assessed paper). The bibliometric snapshot shows extreme topical concentration — the field distribution contains a single domain, fairness-aware model compression and efficient AI hardware, giving a focus score of 100 — and a very compact collaboration footprint of one distinct coauthor, Yiyu Shi (collaboration 14). Output, impact, and activity dimensions cannot yet be computed from the available data. The profile is name-resolved from public bibliographic data and should be read as provisional as the record grows.
Collaboration pattern
The cataloged record shows an exceptionally compact collaboration footprint: exactly one distinct coauthor, Yiyu Shi, who is also the sole frequent coauthor and collaborator on the reviewed 2026 preprint; the collaboration score is 14, near the low end of the scale. Guo sits in the first-author position (non-corresponding) on that paper. With a single-paper basis, this reads as an early, tightly paired working pattern rather than an established network, and should be revisited as more outputs are indexed.
Research keywords
Highlights
Agentic search over compression operators under a fairness constraint
Compression pipeline aimed at FPGA deployment
Medium reproducibility with a narrow empirical base
Timeline
- 2026Posts the first-author arXiv preprint FairCompressAgent on fairness-aware model compression for FPGA deployment.
- 2026Evolve's review credits the framework's search-efficiency gains as credible, while flagging FPGA claims as untested and comparisons based on three runs.
- 2026Reproducibility is assessed at medium readiness, yielding a reproducibility score of 55 over one evaluated paper.
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 scores rest on a single cataloged paper, so collaboration (14), focus (100), and reproducibility (55) are provisional n=1 readings.
- Output, impact, and activity axes have no underlying data in the snapshot and are left unscored.
- Institution and country are unresolved; the profile is name-resolved from public bibliographic data keyed to ORCID 0000-0002-1309-7572.
- The only reviewed item is a 2026 arXiv preprint; no peer-reviewed venue record is yet included.