Katharina von der Wense
University of Colorado Boulder · United States
NLP researcher at CU Boulder focused on topic modeling and multilingual LLM evaluation, with steep recent output growth
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
5
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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 is unscored: this snapshot lacks the complete publication record the axis needs. It would aggregate total works, venue distribution, and per-year output into a productivity score once fuller indexed data becomes available for this researcher.
Impact—
Impact is unscored: citation data in the snapshot is insufficient for a reliable measure. The axis would track citations, h-index trajectory, and downstream influence on other work once a fuller citation record exists.
Focus100/100
Focus scores 100/100: the OpenAlex field distribution lists exactly one field, natural language processing, and topic mentions are tightly concentrated — Topic Modeling (51) and Natural Language Processing Techniques (48) dominate, while domain adaptation and few-shot learning (18), ethics and social impacts of AI (8), and adversarial robustness (8) sit at the margins. The record reads as a single-domain NLP agenda rather than a scattered portfolio.
Collaboration39/100
Collaboration scores 39/100 on a narrow base: the snapshot records only 6 distinct coauthors, and the recurring circle is Mario Sanz-Guerrero, Minh Duc Bui, Abteen Ebrahimi, Sagi Shaier, Peter Herbert Kann, and Manuel Mager. Both Evolve-reviewed papers place her as last author but not corresponding author, a pattern consistent with advisory involvement in a small, stable working group rather than large-scale consortium work.
Activity100/100
Activity scores 100/100 and is the profile's clearest signal: 74 works fall in the most recent three-year window. The counts-by-year series climbs from 3 indexed works in 2023 to 23 in 2024 and 31 in 2025, with 20 already recorded in 2026 (a partial year at snapshot time). This is a steep, sustained productivity ramp.
Reproducibility38/100
Reproducibility scores 38/100 across the 2 Evolve-reviewed papers (reproMean 38). The calibration position paper is rated low readiness — it names no default metric and adds no new evidence — while the cross-lingual medical consistency survey reaches medium readiness, though thin statistics and persona averaging temper its conclusions. Neither paper achieves high readiness in Evolve's assessment.
Research summary
Katharina von der Wense is a natural language processing researcher at the University of Colorado Boulder; this profile is name-resolved from OpenAlex bibliographic data. Her indexed record is exceptionally concentrated: the field distribution lists natural language processing as the only field, and topic mentions are led by Topic Modeling (51) and Natural Language Processing Techniques (48), with smaller clusters in domain adaptation and few-shot learning (18), ethics and social impacts of AI (8), and adversarial robustness in machine learning (8). Activity is the strongest signal — 74 works in the last three years, climbing from 3 indexed works in 2023 to 23 in 2024 and 31 in 2025, with 20 already in 2026. Evolve has reviewed two 2026 arXiv papers carrying her as last author: a position paper arguing that calibration should become a first-class criterion in LLM evaluation, and a stakeholder survey of cross-lingual consistency in LLMs for medical questions. Both place her in a senior, non-corresponding role, and reproducibility readiness across them is low to medium (reproMean 38 over 2 papers). Citation-based output and impact are unscored — the snapshot's h-index of 5 and i10-index of 1 were deemed insufficient — and collaboration remains narrow at six distinct coauthors.
Collaboration pattern
Collaboration is narrow (score 39): the snapshot lists just 6 distinct coauthors, with a stable recurring circle of Mario Sanz-Guerrero, Minh Duc Bui, Abteen Ebrahimi, Sagi Shaier, Peter Herbert Kann, and Manuel Mager. Both catalog papers place her in the last-author, non-corresponding seat, suggesting a senior advisory role within a small working group rather than large multi-institution consortia.
Research keywords
Highlights
Position paper: calibration as a first-class criterion in LLM evaluation
First evidence on cross-lingual consistency of medical LLMs
Timeline
- 2026Last-author position paper 'Calibration as a First-Class Criterion in LLM Evaluation' appears on arXiv.
- 2026Last-author survey 'Perspectives on Cross-Lingual Consistency in LLMs for Medical Questions' appears on arXiv.
- 2026OpenAlex snapshot records 20 indexed works so far in 2026, continuing the ramp from 23 works in 2024 and 31 in 2025.
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.
- Bibliometrics are resolved from OpenAlex only; identity is name-resolved from public bibliographic data, with no ORCID linkage provided.
- Output and impact axes are unscored (insufficient data); the snapshot's h-index of 5 and i10-index of 1 were judged insufficient for a citation-impact score.
- Only 2 papers carry full Evolve review verdicts; the broader record rests on bibliometric metadata without full-text assessment.
- 2026 counts are year-to-date at snapshot time (October 2026), so cross-year output comparisons are partial.