Researcher bridging geometric memory for personalized AI with LLM conversational agents for elderly cognitive care
2
papers here
2
reviewed by Evolve
0
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—
Not yet scored: publication-volume data is unavailable in this snapshot. This axis would measure total research output — papers over time, venues, and authorship roles — across a full bibliographic record rather than the current two-paper 2026 catalog.
Impact—
Not yet scored: no citation counts or attention metrics are present in the data. This axis would measure citations, citation aggregates, and downstream uptake of the work, which require a fuller bibliometric record than the present snapshot provides.
Focus0/100
The focus score is 0, the lowest concentration value on the radar. The raw field distribution splits exactly evenly — 0.5 for machine learning / data mining (personalized AI, user modeling) and 0.5 for natural language processing (conversational AI for elderly healthcare) — across the two 2026 catalog papers, indicating two balanced parallel research tracks rather than one dominant specialty.
Collaboration55/100
Collaboration scores 55. The snapshot records 15 distinct coauthors, with a recurring core of twelve named collaborators, including Jiahong Liu, Menglin Yang, and Wenhao Yu. Across both 2026 papers Irwin King occupies non-first, senior-leaning positions — last author on the geometric-memory paper and middle author on the elderly-care agent paper — a pattern consistent with supervisory or integrative contributions rather than hands-on first-author experimentation.
Activity—
Not yet scored: per-year publication counts are unavailable. This axis would measure recent publishing tempo — outputs per year, recency, and sustained streaks — which cannot be inferred from two same-year 2026 preprints alone.
Reproducibility38/100
Reproducibility averages 38 across the 2 assessed papers (reproMean 38, reproPapers 2). 'Memory Has Geometry' is rated low readiness — a purely conceptual agenda with no experiments — while 'Think Before You Comfort' rates medium, its plausible gains resting on synthetic Cantonese dialogue data and weakly-agreeing LLM-as-judge evaluation. Empirical artifacts currently lag behind the conceptual ambition.
Research summary
Irwin King is a machine learning researcher whose cataloged 2026 work spans two complementary lines: geometric approaches to long-horizon memory for personalized AI, and large language model (LLM) agents for elderly cognitive healthcare. On 'Memory Has Geometry' (arXiv, 2026) the researcher appears as last author, framing long-horizon memory as trajectory-conditioned estimation on non-uniform geometry — a paper Evolve assesses as a compelling but purely conceptual vision with no experiments yet. In the same year, as a middle author on 'Think Before You Comfort', the work contributes Reflective Cognitive Alignment (RCA) for protocol-grounded elderly stimulation agents, coupling structured reasoning with inference-time selection for LLM-based Cognitive Stimulation Therapy agents; its gains are judged plausible but rest on weakly-agreeing LLM judges and synthetic Cantonese dialogue data. The field distribution splits evenly (0.5/0.5) between machine learning / data mining (personalized AI, user modeling) and natural language processing (conversational AI for elderly healthcare), which the focus score of 0 reflects. Collaboration scores 55, with 15 distinct coauthors and a recurring core of twelve named collaborators; reproducibility averages 38 across the two assessed papers. No publication-volume, citation, or per-year activity metrics are available, so output, impact, and activity remain unscored.
Collaboration pattern
The snapshot records 15 distinct coauthors and a stable recurring core of twelve, including Jiahong Liu, Menglin Yang, and Wenhao Yu, yielding a collaboration score of 55. In both 2026 papers Irwin King takes a non-first, senior-leaning position — last author on the geometric-memory vision paper and middle author on the elderly-care agent paper — a pattern consistent with supervisory or integrative roles within a group rather than solo or first-author work.
Research keywords
Highlights
Reflective alignment for protocol-grounded elderly stimulation agents
A geometric agenda for long-horizon personalization memory
Timeline
- 2026Contributes as a middle author to 'Think Before You Comfort', introducing Reflective Cognitive Alignment for protocol-grounded LLM elderly stimulation agents.
- 2026Appears as last author on 'Memory Has Geometry', laying out a conceptual, experiment-free agenda for non-uniform geometric memory in long-horizon personalized AI.
- 2026Both cataloged outputs land as arXiv preprints in the same year, splitting the field distribution evenly between machine learning / data mining and natural language processing.
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 signals come from Evolve's catalog of two 2026 arXiv preprints; no citation or venue data beyond these is included.
- Institution, country, and ORCID are not provided; this profile is name-resolved from public bibliographic data only.
- Output, impact, and activity axes are unscored because publication-volume and citation metrics are unavailable in this snapshot.
- The reproducibility score (mean 38) rests on only 2 assessed papers and is sensitive to this small sample.