Vision and autonomous-driving researcher working on incremental object detection and end-to-end driving
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-14) 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 publication-count or productivity metrics are available in this snapshot, so the output axis is unscored. It would measure the volume of research outputs — papers, preprints and venues — over time.
Impact—
Citation-based metrics are not yet available for this profile, so the impact axis is unscored. It would measure scholarly influence via citations, citing works, and uptake of the researcher's methods and benchmarks.
Focus0/100
The focus score is 0/100, reflecting an even 50/50 split between two fields: autonomous driving / robot learning (embodied AI) and computer vision (incremental learning for object detection). The two reviewed papers sit one in each field, so this reads as a deliberately dual-track profile rather than a narrow specialist one.
Collaboration51/100
The collaboration score is 51/100, based on 12 distinct coauthors in the underlying data. The reviewed work bridges two communities: vision-oriented coauthors such as Konrad Schindler, Luc Van Gool and Danda Pani Paudel, and driving-oriented ones including Marco Pavone, Boris Ivanovic, Kashyap Chitta, Maximilian Igl and Peter Karkus. On both cataloged 2026 papers Sakaridis appears in the last author position.
Activity—
Recent publication-rate data is insufficient in this snapshot, so the activity axis is unscored. It would measure research cadence — how regularly new outputs appear in recent years.
Reproducibility55/100
The reproducibility score is 55/100, computed as the mean readiness across 2 reviewed papers (reproMean 55, reproPapers 2). Both carry a medium reproducibility rating: MECAIL reports controlled comparisons on D-RICO and ODinW-13, while OPTED is described as rigorous and well-controlled, though the review flags minor concerns on collision metrics.
Research summary
Christos Sakaridis (ORCID 0000-0003-1127-8887) works across two adjacent fields: autonomous driving / robot learning (embodied AI) and computer vision, specifically incremental learning for object detection. The Evolve catalog currently contains two of his 2026 arXiv papers, on both of which he holds the last author position (neither is marked corresponding). OPTED introduces on-policy fine-tuning for end-to-end driving with a render-free reinforcement-learning teacher; Evolve's review credits it with raising closed-loop driving scores by 1.6x and 9.5x, calling it rigorous and well-controlled while noting minor concerns on collision metrics. MECAIL proposes communication-aware incremental object detection with compact 14.6 KB spatiotemporal experts; the review reports that it matches baselines on the D-RICO benchmark while lagging on ODinW-13 under full training. His collaboration profile is moderate (51/100) with 12 distinct coauthors, bridging a vision-oriented cluster (Konrad Schindler, Luc Van Gool, Danda Pani Paudel) and a driving-oriented cluster (Marco Pavone, Boris Ivanovic, Kashyap Chitta, Maximilian Igl, Peter Karkus). An even 50/50 field split yields a low focus score (0/100), indicating a deliberate dual-track profile rather than a narrow specialty. Reproducibility readiness is rated medium for both reviewed papers, with a mean of 55/100 across 2 papers. The output, impact and activity axes remain unscored because their underlying metrics are not yet available.
Collaboration pattern
The data lists 12 distinct coauthors and a moderate collaboration score of 51/100. The reviewed work bridges two clusters: a computer-vision cluster (Konrad Schindler, Luc Van Gool, Danda Pani Paudel, Damiano Da Col) and an autonomous-driving / embodied-AI cluster (Marco Pavone, Boris Ivanovic, Kashyap Chitta, Maximilian Igl, Peter Karkus, Matthias Neuwirth-Trapp, Maarten Bieshaar). On both cataloged 2026 papers Sakaridis is the last author, though neither is marked corresponding.
Research keywords
Highlights
Compact 14.6 KB experts for communication-constrained incremental object detection
Render-free RL teacher lifts closed-loop driving scores by 1.6x and 9.5x
Dual-track 2026 output bridging vision and driving
Both reviewed papers appeared in 2026 with Sakaridis in the last author position, one in autonomous driving / robot learning (embodied AI) and one in computer vision (incremental object detection) — a 50/50 field split that defines this profile's shape. Mean reproducibility readiness across the two papers is 55/100.
Timeline
- 2026OPTED — on-policy fine-tuning for end-to-end driving with a render-free teacher — posted to arXiv; Sakaridis in the last author position.
- 2026MECAIL — communication-aware incremental learning for object detection with 14.6 KB spatiotemporal experts — posted to arXiv; last author position.
- 2026The reviewed collaboration network for this period spans 12 distinct coauthors across vision and driving communities; mean reproducibility readiness across the 2 reviewed papers is 55/100.
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.
- Identity is name-resolved from public bibliographic data under ORCID 0000-0003-1127-8887; institution and country are not present in this snapshot.
- Dimension scores come from a limited bibliometric snapshot; the output, impact and activity axes lack sufficient data and are unscored.
- Reviewed-paper evidence covers only 2 papers from the Evolve catalog, both 2026 arXiv preprints; peer-review status and the full publication record are not represented.
- Field distribution reflects only the two reviewed-paper fields (50/50) and may under-represent the researcher's actual breadth.