Machine-learning researcher in adaptive computer vision, few-shot meta-learning, and vision-based plant phenotyping.
3
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
21
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—
A composite output score is not yet available for this profile: the underlying record lacks the complete work-level data needed to score productivity, though raw indicators (h-index 21, i10-index 36) are on file.
Impact—
No impact score is available: citation-level data for this profile's works is insufficient in the current snapshot, so influence — citations and per-paper impact — cannot yet be scored; only raw h-index 21 and i10-index 36 are recorded.
Focus100/100
The focus dimension scores 100/100. Output concentrates tightly on machine-learning perception: the five recorded topic clusters all sit in one family — context-aware activity recognition (31 works), advanced neural network applications (21), emotion and mood recognition (18), domain adaptation and few-shot learning (16), and time series analysis and forecasting (15) — with the catalog field listed as computer vision for agricultural phenotyping.
Collaboration50/100
Collaboration scores 50/100. Over the last three years this profile records 59 works with 11 distinct co-authors — a modest pool relative to output volume, suggesting work spread across several small teams rather than one large consortium. Recurring partners include Muhammad Salman Siddiqui, Chandler Timm C. Doloriel, Yunbei Zhang, Kristian Hovde Liland, and Habib Ullah.
Activity100/100
Activity scores 100/100. Annual work counts climb steadily through the 2020s — from 2 works in each of 2019 and 2020 to 7 in 2021, 14 in 2023, 15 in 2024, 21 in 2025, and 23 in 2026 — with 59 works in the most recent three-year window, the strongest sustained output on record for this profile.
Reproducibility55/100
Reproducibility scores 55/100 (reproMean 55), based on one Evolve-reviewed paper so far. That 2026 meta-learning plant growth study earned a medium repro-readiness rating, with honest ablations and solid statistics offset by a flagged random-split leakage risk. The single-paper basis makes this an early-stage, low-confidence estimate.
Research summary
Fadi Al Machot is a machine-learning researcher at the Norwegian University of Life Sciences whose record, name-resolved from public OpenAlex bibliographic data, spans adaptive computer vision and applied deep learning. The snapshot reports an h-index of 21 and an i10-index of 36, with annual counts summing to 137 works across 2010–2026 and a clear late-career acceleration: output rises from 2 works in each of 2019 and 2020 to 14 in 2023, 15 in 2024, 21 in 2025, and 23 in 2026, with 59 works in the most recent three-year window. Topic clustering confirms a focused agenda: context-aware activity recognition systems (31 works), advanced neural network applications (21), emotion and mood recognition (18), domain adaptation and few-shot learning (16), and time series analysis and forecasting (15). The 2026 catalog entries show the current direction — continual test-time adaptation of vision-language models for open-vocabulary segmentation, entropy sensitivity-guidance in a strict online setting, and a data-efficient plant growth estimation approach pairing MAML++ meta-learning with vision transformers and fuzzy c-means task construction, on which Al Machot is last author. Evolve's review of that paper credits honest ablations and solid statistics while flagging random-split leakage risk and modest gains, giving a medium repro-readiness rating and a reproducibility score of 55 drawn from a single assessed paper. Collaboration is moderate, with 11 distinct co-authors across the recent output.
Collaboration pattern
Collaboration is moderate (score 50): the last three years show 59 works with only 11 distinct co-authors — a compact pool for that output volume. Frequent partners include Muhammad Salman Siddiqui, Chandler Timm C. Doloriel, Yunbei Zhang, Habib Ullah, Kristian Hovde Liland, Tor Kristian Stevik, Sheikh Hasan Elahi, Rakibul Islam, and Sarthak Kumar Maharana, pointing to small-team co-authorship on the recent adaptation and phenotyping work rather than large-consortium projects.
Research keywords
Highlights
Few-Shot Meta-Learning for Vision-Based Plant Growth Estimation
Continual Test-Time Adaptation of Vision-Language Models in Open-Vocabulary Segmentation
Co-authored 2026 preprint extending continual test-time adaptation to vision-language models in open-vocabulary semantic segmentation, targeting adaptive inference under distribution shift. It sits at the core of the recent adaptation-focused output but is not yet reviewed by Evolve (no verdict or repro-readiness rating on file).
Entropy Sensitivity-Guidance for Strictly Online Continual Test-Time Adaptation
2026 preprint proposing an entropy sensitivity-guidance scheme for continual test-time adaptation in a strict online setting, complementing the open-vocabulary segmentation work. Like its companion paper, it awaits Evolve review, so quality and reproducibility remain unrated in the catalog.
Timeline
- 2010Earliest works appear in the annual counts, with 3 works recorded.
- 201612 works recorded, the strongest year of the 2010s in the annual counts.
- 2021Output recovers to 7 works after the 2019–2020 low of 2 works per year.
- 2023A steep climb begins: 14 works, rising to 15 in 2024 and 21 in 2025.
- 2026Three catalog papers appear — two continual test-time adaptation preprints and a meta-learning plant growth study — within a record 23 works for the year.
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 indicators are drawn from an OpenAlex snapshot only (identity: openalex); no other databases were consulted.
- Composite scores for the output and impact dimensions are unavailable; only the raw h-index (21) and i10-index (36) are on record, and no per-paper citation counts were provided.
- The reproducibility score (55) rests on a single Evolve-reviewed paper; the other two 2026 catalog papers have no verdict, keywords, or repro-readiness rating yet.
- This profile is name-resolved from public bibliographic data; the source lists the affiliation country as DE while the institution is the Norwegian University of Life Sciences, so biographical details should be treated with caution.