Jordy Kieto

—

Unclaimed · AI-structured profileOpenAlex

Early-stage researcher in imitation learning for autonomous driving; first & corresponding author of a 2026 CARLA study

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

—

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-09-21) 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—

Insufficient data: the record holds a single reviewed arXiv preprint, so no publication-count metric can be computed. This axis would measure cumulative papers, venue spread, and first/corresponding-author volume over time.

Impact—

Insufficient data: this snapshot contains no citation counts, h-index, or adoption signals. The axis would measure how often the CARLA behavioral-cloning work is cited, reused, or built upon by other researchers.

Focus100/100

Scores 100 from a perfectly concentrated field distribution: the single catalogued paper sits entirely in robotics / machine learning (autonomous driving). A one-paper record makes concentration trivially maximal, so genuine focus should be re-assessed as outputs accumulate. Topic keywords — imitation learning, behavioral cloning, closed-loop evaluation, covariate shift — reinforce the single-theme profile.

Collaboration0/100

Scores 0 because the bibliometric record lists 0 distinct coauthors and no frequent collaborators; the single catalogued paper is first-authored and corresponding-authored by Kieto, consistent with solo-led work. The dataset is one paper deep, so this score reflects an empty collaboration ledger rather than an established pattern; joint publications would move it quickly.

Activity—

Insufficient data: only one paper, dated 2026, with no multi-year publication series. This axis would measure publishing cadence and continuity of research activity across recent years.

Reproducibility100/100

Scores 100 as the mean reproducibility readiness across exactly 1 assessed paper. Evolve's review of the CARLA imitation-learning preprint found strong artifacts and notable research honesty, assigning 'high' reproducibility readiness; however, the headline closed-loop claims remain unmeasured, and a single-paper sample is too thin to read as a stable estimate.

Research summary

Jordy Kieto is an early-stage researcher working at the intersection of robotics and machine learning, focused on imitation learning for autonomous driving. Evolve's catalog contains one reviewed paper — a 2026 arXiv preprint, 'Imitation Learning for Autonomous Driving in CARLA' — on which Kieto is first author and corresponding author. The paper claims that a compact 1.36M-parameter behavioral-cloning policy can drive for hours without collisions in the CARLA simulator. Evolve's verdict credits the work with strong artifacts and unusual research honesty, assigning 'high' reproducibility readiness, while cautioning that the headline closed-loop performance claims remain unmeasured — the central open question for follow-up work. The bibliometric snapshot is correspondingly thin: a single output in one field, robotics / machine learning (autonomous driving), zero distinct coauthors, and no frequent collaborators, which yields a maximal focus score and a maximal — but single-paper — reproducibility score of 100. Output, impact, and activity axes cannot be scored yet. No institution, country, or ORCID is attached to the record, so this profile is name-resolved from public bibliographic data. As a portrait, it is best read as a strong, verifiable opening move — an artifact-heavy, honest evaluation study in simulation — rather than an established body of work.

Collaboration pattern

The available record lists zero distinct coauthors and no frequent collaborators, and the single catalogued paper is both first-authored and corresponding-authored by Kieto — a pattern consistent with independent, solo-led work. Because the evidence is one paper deep, this is an empty ledger rather than a demonstrated preference; any future co-authorships are not yet visible in the data.

Research keywords

imitation learningbehavioral cloningautonomous drivingCARLA simulatorclosed-loop evaluationcovariate shiftroboticsmachine learningend-to-end driving policiesreproducible research artifacts

Highlights

First-authored CARLA imitation-learning study (2026)
Kieto is first and corresponding author of 'Imitation Learning for Autonomous Driving in CARLA', a 2026 arXiv preprint claiming a 1.36M-parameter behavioral-cloning policy drives hours collision-free in the CARLA simulator.
High reproducibility readiness with strong artifacts
Evolve's review found the paper's artifacts and honesty notably strong, assigning 'high' reproducibility readiness — the basis of the reproducibility score of 100 (reproMean across 1 assessed paper).
Open question: headline closed-loop claims unmeasured
The same review cautions that the headline closed-loop performance claims remain unmeasured — the clearest gap for follow-up work and for anyone building on this policy.

Timeline

2026Published 'Imitation Learning for Autonomous Driving in CARLA' on arXiv as first author and corresponding author.
2026Evolve reviewed the paper, crediting strong artifacts and honesty with 'high' reproducibility readiness.
2026The review flagged the headline closed-loop collision-free claims as promising but still unmeasured.

Papers in the Evolve catalog

2026-09-17first authorcorresponding
Imitation Learning for Autonomous Driving in CARLA

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.

No institution, country, or ORCID is attached to this record; the profile is name-resolved from public bibliographic data and name ambiguity is possible.
All bibliometric scores rest on a single reviewed arXiv preprint; output, impact, and activity are unscored for lack of data.
The reproducibility score of 100 is a mean over exactly one assessed paper (reproPapers = 1) and should not be read as a stable estimate.
The collaboration score of 0 reflects zero distinct coauthors in the available snapshot, not an assessment of teamwork quality.