HCI researcher at the University of Washington studying design guidance and generative AI in human-AI interaction
2
papers here
1
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) 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 data is not yet available in this snapshot, so no production score is shown. This axis would measure publication volume and consistency over time, including annual output counts and authorship roles across the record.
Impact—
Impact metrics are missing from the current data, so no citation-based score is reported. This axis would capture how widely the work is cited and taken up, including citation counts, aggregate indices, and influence across venues.
Focus100/100
Focus scores 100 because the field-tagged catalog is fully concentrated: the field distribution lists human-computer interaction as the only field with one tagged paper. The first-authored AI companion study and the thematically adjacent generative AI paper on student communication indicate a coherent human-AI interaction agenda rather than scattered topics.
Collaboration36/100
Collaboration scores 36, based on 5 distinct coauthors in the current snapshot. Frequent coauthors include Deveshi Modi, Divya Mavinkurve, Jieqiong Ding, Mark Zachry, and Jared Ren. The first-authored multi-method study implies a multi-rater team, with the reviewed verdict noting 3-4 raters per product, but the small distinct-coauthor count indicates a compact, concentrated network so far.
Activity—
Activity data is insufficient in this snapshot, so no cadence score is shown. This axis would measure recent publishing rhythm, including time since the latest paper, per-year submission counts, and sustained output across recent years.
Reproducibility20/100
Reproducibility scores 20, computed from a single assessed paper (reproPapers: 1, mean readiness 20). The reviewed multi-method study is rated low readiness: Evolve's verdict states the process is well documented but the statistics rest on only 3-4 raters per product. With one paper assessed, this score reads as an early, narrow signal rather than a stable trait.
Research summary
Soobin Cho is a University of Washington-affiliated researcher working in human-computer interaction, focused on how people design and interact with AI-driven interfaces. The Evolve catalog lists two 2026 arXiv entries. The first, a first-authored multi-method study titled 'Assessing the Applicability of Existing Design Recommendations to AI Companion Design', asks whether established design recommendations apply to AI companions; Evolve's verdict is that existing design guidance transfers poorly to AI companions, with a well-documented process but statistics resting on only 3-4 raters per product. The second, 'Delegating Before Learning: Where Generative AI Sits in Students' Professional Communication', lists Cho as last author and examines generative AI's place in students' professional communication; it has not yet received an Evolve review. The bibliometric snapshot yields a focus score of 100, with human-computer interaction as the only tagged field; a collaboration score of 36 from 5 distinct coauthors; and a reproducibility score of 20 computed from one assessed paper. Frequent coauthors include Deveshi Modi, Divya Mavinkurve, Jieqiong Ding, Mark Zachry, and Jared Ren. Output, impact, and activity dimensions currently lack sufficient data and are left unscored. This is an early-stage, HCI-centered profile: the visible record shows an emerging human-AI interaction agenda spanning design-method evaluation and AI use in communication practice, supported by a compact collaboration network and limited public metrics.
Collaboration pattern
The snapshot lists 5 distinct coauthors, with Deveshi Modi, Divya Mavinkurve, Jieqiong Ding, Mark Zachry, and Jared Ren appearing as frequent collaborators. The pattern suggests a small, stable team at the University of Washington: Cho leads a multi-method HCI evaluation as first author while also contributing as last author to a separate study on generative AI in student communication. The collaboration score of 36 reflects this compact network rather than a broad set of partners.
Research keywords
Highlights
First-authored test of whether existing design guidance fits AI companions
Last-author study on generative AI in students' professional communication
'Delegating Before Learning: Where Generative AI Sits in Students' Professional Communication' (2026, arXiv) lists Soobin Cho as last author. It examines where generative AI sits in students' professional communication; the catalog record carries no Evolve verdict, keywords, or reproducibility rating yet.
A concentrated human-AI interaction agenda in the 2026 catalog
Both catalog entries are dated 2026 and sit in adjacent territory — design methods for AI companions and generative AI in student communication — consistent with the focus score of 100, where human-computer interaction is the only field in the field distribution.
Timeline
- 2026First-authored arXiv preprint assessing whether existing design recommendations apply to AI companion design; the Evolve review notes a well-documented process but only 3-4 raters per product.
- 2026Last-authored arXiv preprint 'Delegating Before Learning' on where generative AI sits in students' professional communication; not yet reviewed by Evolve.
- 2026Both papers currently visible in the Evolve catalog are dated 2026, so the provided record spans a single year and contains no earlier timeline points.
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.
- Name is common; the profile is name-resolved from public bibliographic data and no ORCID is provided, so attribution to the specific individual should be verified.
- Bibliometric scores come from OpenAlex-based public bibliometric data only; venue acceptance, citation records, and career information are out of scope.
- Output, impact, and activity axes lack sufficient data and are reported as missing rather than scored.
- The reproducibility score of 20 rests on a single assessed paper (reproPapers: 1), so it is an early, narrow signal.