Yonsei researcher in LLM-based time series forecasting; two 2026 preprints, strong reproducibility signals.
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-09) 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 scored: OpenAlex reports h-index 0 and i10-index 0 with only 1 indexed work in the recent three-year window, so publication volume cannot be assessed. This axis would measure output volume, venues, and authorship position.
Impact—
Not scored: OpenAlex shows h-index 0 and i10-index 0, and the two 2026 preprints are too recent for citation accumulation. This axis would measure citation uptake, citing works, and per-paper attention once citations accrue.
Focus0/100
The focus dimension scores 0/100. The OpenAlex field distribution splits evenly — 0.5 "machine learning (LLM-based time series forecasting)" and 0.5 "machine learning (time series forecasting)" — and the indexed topics (topic modeling; forecasting techniques and applications; time series analysis and forecasting) all point to one narrow area. The low score reflects the tiny indexed sample rather than scattered interests; both catalog papers target the same problem.
Collaboration39/100
Scores 39/100. OpenAlex records 6 distinct coauthors; the catalog's recurring collaborators are Daeun Ji, Minkyoung Kim, Beakcheol Jang, Dongkuk Kim and Beomsoo Kim. Both 2026 papers list Yohan Lee as a middle author without corresponding status, a pattern consistent with contributing inside a small, stable Yonsei-affiliated team rather than leading projects or sustaining a large cross-institution network.
Activity20/100
Scores 20/100. OpenAlex countsByYear lists a single work in 2026 and recent3yWorks equals 1, though the Evolve catalog already holds two 2026 arXiv preprints — a sign that indexing lags. The record is brand new, so activity sits at a starting baseline rather than showing decline.
Reproducibility78/100
Scores 78/100, the mean readiness across 2 assessed papers (reproPapers: 2). "Beyond Similarity" rates high — Evolve calls its statistics unusually honest — while "CTRL" rates medium, flagged for missing error bars and a Table 3 mismatch. The signal suggests above-average attention to reproducibility, with rigor gaps concentrated in statistical reporting.
Research summary
Yohan Lee is an early-stage machine learning researcher at Yonsei University (South Korea) whose visible record concentrates on time series forecasting with large language models. The Evolve catalog holds two 2026 arXiv preprints, both with Lee in a middle authorship position and neither as corresponding author. The first, CTRL, casts LLM agents as label-free residual controllers for frozen forecasting backbones; Evolve's review finds the gains track distribution shift but notes missing error bars and a Table 3 mismatch. The second, Beyond Similarity, introduces a parameter-free coverage regularizer for prompt retrieval in LLM-based forecasting; Evolve highlights unusually honest statistics while noting the aggregate gain resolves on only 1 of 24 cells. OpenAlex bibliometrics are thin: h-index 0, i10-index 0, one indexed work in the recent three-year window, and six distinct coauthors, with recurring collaborators including Daeun Ji, Minkyoung Kim, Beakcheol Jang, Dongkuk Kim and Beomsoo Kim. The field profile splits evenly between LLM-based time series forecasting and time series forecasting. The strongest measurable signal is reproducibility: a mean readiness of 78 across the two assessed papers, one rated high and one medium. This is an early-stage profile built from a small sample; citation-based output and impact cannot yet be scored.
Collaboration pattern
OpenAlex records 6 distinct coauthors. Recurring collaborators in the catalog are Daeun Ji, Minkyoung Kim, Beakcheol Jang, Dongkuk Kim and Beomsoo Kim (listed twice in the source data, likely a duplicate). The collaboration dimension scores 39/100. On both 2026 papers Yohan Lee is a middle author without corresponding status, indicating contribution within a small, stable Yonsei-affiliated team rather than an independent or large cross-institution network; no international partners are visible in this data.
Research keywords
Highlights
CTRL: LLM agents as residual controllers for frozen forecasting models
Beyond Similarity: coverage-aware prompt selection for LLM forecasting
Timeline
- 2026Enters the Evolve catalog with CTRL (arXiv), a control-based time series forecasting method using LLM-guided residual learning; middle authorship, non-corresponding.
- 2026Publishes Beyond Similarity (arXiv), a coverage-aware prompt-selection method for LLM-based forecasting; Evolve rates its reproducibility readiness high.
- 2026A reproducibility profile takes shape: 2 assessed papers with a mean readiness score of 78, the strongest measured dimension in this early record.
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 come from OpenAlex only; no ORCID is linked, so name resolution may miss works outside this identity.
- OpenAlex indexes only 1 work in the recent three-year window while the Evolve catalog holds 2 papers from 2026, so coverage likely lags the actual record.
- Both reviewed papers are 2026 arXiv preprints without peer-review status; citation impact cannot yet be assessed (h-index 0, i10-index 0).
- The profile rests on only 2 reviewed papers and a thin OpenAlex record; dimension scores may shift substantially as indexing catches up.