University of Pittsburgh researcher benchmarking LLM-generated feedback and pedagogical alignment in NLP for education.
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
1
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 metrics are not available in this snapshot: the OpenAlex record behind this profile lacks a reliable publication count or productivity measure, so this axis would measure total works and yearly output once the identity record is resolved.
Impact—
Impact data is insufficient: the underlying record shows an h-index of 1 and i10-index of 0, which cannot support a meaningful citation-impact score for a full career; this axis would measure citations and citation-based influence once resolved.
Focus100/100
This profile scores 100 for focus, the maximum. The topic distribution is tightly concentrated: Intelligent Tutoring Systems and Adaptive Learning leads with 7 works, followed by Topic Modeling with 5 and Natural Language Processing Techniques with 4, plus Innovative Teaching and Learning Methods and Student Assessment and Feedback at 3 each, and a single registered field, NLP for education and learning analytics.
Collaboration41/100
Collaboration scores 41 on a 0-100 scale. The recent record shows 9 works in the last three years with 7 distinct coauthors, including repeated collaborators Mohamed Elaraby, Ahmed Elhady, Norah Almousa, Shayan Peyghambari Oskoui, Raquel Coelho, Gayle Rogers and Xiang Lorraine Li. Both catalogued 2026 papers place Litman in the last author position, a pattern typically associated with senior supervisory authorship in small teams.
Activity67/100
Activity scores 67. The yearly counts show current momentum: 8 works are dated 2026, and the record includes 9 works in the most recent three-year window. Earlier points are sparse — 1 work in 2024, 1 in 2021, and 3 in 2013 — so the profile reads as recently active, with the current burst concentrated in 2026.
Reproducibility55/100
Reproducibility scores 55, based on a single Evolve-reviewed paper (reproMean 55 across 1 assessed work). That paper — the LLM-generated feedback benchmark — carries a medium repro-readiness rating. With only one assessed paper, this score is a narrow signal and should firm up as more of the catalogue is reviewed.
Research summary
Diane Litman is a researcher at the University of Pittsburgh (US) whose recorded work sits at the intersection of natural language processing and education. The OpenAlex topic profile is tightly concentrated: Intelligent Tutoring Systems and Adaptive Learning leads with 7 works, followed by Topic Modeling with 5, Natural Language Processing Techniques with 4, and Innovative Teaching and Learning Methods and Student Assessment and Feedback with 3 each. Current activity is concentrated in 2026, with 8 works dated that year and 9 works across the most recent three-year window. Two 2026 arXiv papers are catalogued by Evolve, both with Litman as last author: a benchmark study of feedback focus and pedagogical adaptivity in LLM-generated feedback on student writing — Evolve's review finds LLM feedback covers most focus types but diverges from the teacher distribution, with partial adaptivity and underpowered statistics — and a sequence-level distillation study improving argument saliency coverage in small LLMs for long legal opinion summarization. The collaboration footprint lists 7 distinct coauthors in the recent window. Quantitative impact and output scores are not yet computable because the underlying record reports only an h-index of 1 and an i10-index of 0, an unusually sparse signature; this profile should therefore be read as an early-stage, name-resolved bibliographic view.
Collaboration pattern
The record shows 9 works over the last three years with 7 distinct coauthors. Recurring names include Mohamed Elaraby, Ahmed Elhady, Norah Almousa, Shayan Peyghambari Oskoui, Raquel Coelho, Gayle Rogers and Xiang Lorraine Li. On both 2026 papers catalogued by Evolve, Litman appears as last author (non-corresponding), a position typically associated with senior supervision of student- or junior-led projects, while day-to-day authorship sits with earlier-career collaborators.
Research keywords
Highlights
New FeedType benchmark: LLM feedback diverges from teacher distribution
Sequence-level distillation boosts argument saliency in legal summarization
A second 2026 arXiv paper, last-authored by Litman, improves argument saliency coverage in small LLMs for long legal opinion summarization via sequence-level distillation. It extends the profile's recorded focus beyond education into legal NLP applications; Evolve has not yet reviewed it, so no verdict or repro-readiness rating exists.
Output surge: 8 works dated 2026
The bibliometric snapshot logs 8 works dated 2026 — the highest single-year count on record, versus 3 in 2013, 2 in 2009 and 1 each in 2012, 2021 and 2024 — and 9 works in the most recent three-year window, marking the profile's most active period.
Timeline
- 2009Earliest recorded activity in the snapshot: 2 works dated 2009.
- 2012–20131 work in 2012 followed by 3 in 2013 — the largest pre-2026 annual count in the record.
- 2021Sparse interim activity: 1 work recorded in 2021.
- 2024Activity resumes: 1 work recorded in 2024, ahead of the 2026 surge.
- 2026Peak year: 8 works dated 2026, including two arXiv papers catalogued by Evolve with Litman as last author — the LLM feedback benchmark and the legal summarization distillation study.
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.
- All bibliometrics come from a single OpenAlex record whose citation fields (h-index 1, i10-index 0) are unusually sparse and may reflect an incomplete identity match rather than full career output.
- Output and impact dimensions are unscored (dimMissing): the snapshot reports no reliable publication totals or citation aggregates.
- Only 2 papers (both 2026 arXiv preprints) are in the Evolve catalog and only 1 has been reviewed; reproducibility rests on a single assessed paper (reproMean 55, reproPapers 1).
- Yearly counts exist only for 2009, 2012, 2013, 2021, 2024 and 2026; intervening years are absent, so longitudinal trends are partial.