University of Toronto IR & scholarly NLP researcher: LLM-judge evaluation, peer-review analytics, agentic deep search
3
papers here
3
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
31
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—
No output score is available: the snapshot lacks the aggregates this axis needs, such as total and venue-normalized publication counts. It would measure overall research production volume; the data is insufficient, so the axis is left unscored rather than estimated.
Impact—
No impact score is available: citation totals and normalized citation metrics were not provided. This axis would measure citation-based influence of the work; the snapshot lists only h-index 31 and i10-index 112, so it stays unscored.
Focus0/100
The focus score is 0, reflecting a dispersed topic profile rather than a narrow specialty. The topic record spreads across topic modeling (118 works), recommender systems (103), information retrieval and search behavior (103), and software engineering methodologies and practices (54 and 52), while the current field distribution splits evenly — one third each — across LLM-based evaluation, scholarly NLP, and agentic deep search.
Collaboration51/100
Collaboration scores 51 of 100. The last three years show 12 distinct co-authors, and the frequent-coauthor list pairs senior researchers such as Charles L. A. Clarke and Benjamin C. M. Fung with a recurring group that includes Sajad Ebrahimi, Negar Arabzadeh, and Shakiba Amirshahi. All three reviewed papers list Bagheri in the last author position, consistent with a senior role on recent work.
Activity100/100
Activity scores a perfect 100. The counts-by-year record shows publications in every year from 2004 to 2026, with 87 works in the last three years. Recent annual counts climb from 17 (2024) to 27 (2025) to 43 (2026) — the highest in the series — alongside three 2026 arXiv preprints reviewed by Evolve.
Reproducibility70/100
Reproducibility averages 70 across the three Evolve-reviewed papers. Peerify is rated high readiness (an 800-claim benchmark with well-audited supervision), while the LLM-judge reliability analysis and Question's Gambit are rated medium — the former constrained by nested data and data-derived equivalence bounds, the latter by a missing component ablation.
Research summary
Ebrahim Bagheri is a researcher at the University of Toronto whose indexed work spans information retrieval, natural language processing, and software engineering. The OpenAlex record shows more than two decades of continuous output (h-index 31, i10-index 112), with publications recorded every year from 2004 through 2026 and 87 works in the last three years — an activity level the radar scores at 100. Topically the record is broad: topic modeling (118 works), recommender systems (103), information retrieval and search behavior (103), and software engineering methodologies and practices (54 and 52), and the focus score is 0, with the current field distribution splitting evenly across LLM-based evaluation and scientometrics, scholarly NLP, and agentic deep search. Three 2026 arXiv papers reviewed by Evolve, all listing Bagheri as last author, target peer-review analytics and retrieval: a reliability analysis finding 23 of 29 LLM-judge review metrics conflate style with substance, the Peerify claim-verification benchmark (800 claims, six variants), and Question's Gambit, which lifts Pi-Serini accuracy on BrowseComp-Plus from 83.1 to 90.5. Reproducibility readiness averages 70 across the three, and collaboration scores 51 with 12 distinct co-authors over the last three years.
Collaboration pattern
Recent collaboration is compact and steady: 12 distinct co-authors over the last three years, giving a radar collaboration score of 51. The frequent-coauthor list pairs senior researchers such as Charles L. A. Clarke and Benjamin C. M. Fung with a recurring group including Sajad Ebrahimi, Negar Arabzadeh, and Shakiba Amirshahi. All three 2026 catalog papers place Bagheri in the last author position, a slot conventionally tied to supervisory roles.
Research keywords
Highlights
Auditing LLM-judge metrics for peer review
Peerify: verifying peer-review claims at scale
First-move question decomposition for agentic search
Timeline
- 2004Earliest publication record in the OpenAlex counts-by-year series (1 work).
- 2017Strongest year before the 2020s: 31 publications recorded.
- 2025Annual count reaches 27, continuing an upward trend from 17 in 2024.
- 2026Highest annual count in the record: 43 publications, year-to-date.
- 2026Three arXiv preprints reviewed by Evolve — Peerify, Question's Gambit, and the LLM-judge reliability analysis — all with Bagheri as last author.
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 drawn from OpenAlex only (identity: openalex); the output and impact axes lack their underlying aggregates and are left unscored.
- Identity is name-resolved from public bibliographic data (ORCID 0000-0002-5148-6237); affiliation is as listed in the source record.
- Evolve verdicts cover only three 2026 arXiv preprints and may not represent the full publication record.
- 2026 counts are year-to-date and may rise before the year closes.