Bolin Gao

Tsinghua University

Unclaimed · AI-structured profileOpenAlex

Tsinghua researcher advancing risk-aware occupancy for safety-oriented end-to-end autonomous driving

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

—

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-15) 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—

The output axis is not scored yet: publication counts and productivity metrics for this researcher are absent from the current snapshot. It would measure total and per-year paper volume across journals, conferences, and preprint servers.

Impact—

The impact axis is not scored yet: no citation counts, h-index, or per-paper influence data are available. It would measure how widely the two risk-aware occupancy papers and any prior work are cited and taken up downstream.

Focus0/100

The radar records a focus score of 0, the dataset's floor, with the raw field distribution split exactly 50/50 between autonomous driving/computer vision and autonomous driving/machine learning. In substance the agenda is narrow: both cataloged 2026 papers target risk-aware occupancy for end-to-end driving. Because the two adjacent field labels each take half, no single-field concentration registers, and the axis scores at the bottom of the scale.

Collaboration39/100

The collaboration score is 39, on the lower end, drawn from 6 distinct coauthors across the two cataloged papers. Recurring partners include Jiaxing Chen, Hengduo Zou, Yiren Zhao, YuKai Qin, and Lidong Yu. Bolin Gao sits in the last author position on both preprints without a corresponding-author flag, a pattern consistent with a small, stable, single-group team rather than a wide external network.

Activity—

The activity axis is not scored yet: the snapshot lacks recent-publication cadence and year-by-year output data. It would measure publishing tempo and continuity over time, here only inferable from two 2026 arXiv entries.

Reproducibility55/100

Reproducibility scores 55, the mean across 2 evaluated papers, both rated medium readiness. The evidence base is thin but consistent: benchmark-anchored evaluation on NAVSIM (PDMS 0.937→0.951), nuScenes, and GenAD supports replication in principle, while Evolve's review notes reliance on oracle future-derived risk labels and statistically untested small-count collision metrics as caveats.

Research summary

Bolin Gao is a researcher at Tsinghua University working on safety-oriented end-to-end autonomous driving, with a cataloged focus on risk-aware occupancy — representations that let a driving policy reason about hazard before committing to trajectories. Evolve's catalog holds two 2026 arXiv papers, both listing Gao in the last (senior) author position, neither flagged as corresponding author. The first, 'Risk-Aware Occupancy for Safety-Oriented End-to-End Autonomous Driving,' evaluates on nuScenes and reports GenAD collision-rate drops of 35–53%, which Evolve judges promising while cautioning that small-count open-loop metrics lack statistical testing. The second, 'Driving on Registers, Reasoning on Risk,' adds risk-aware occupancy to a NAVSIM register-based planner, lifting PDMS from 0.937 to 0.951; Evolve notes the headline gains rely on oracle future-derived risk labels. Both papers are rated medium reproducibility readiness, giving a reproducibility score of 55 (mean over 2 papers). The field distribution splits evenly between autonomous driving/computer vision and autonomous driving/machine learning, and the recorded focus score is 0. Collaboration is compact: 6 distinct coauthors, with Jiaxing Chen, Hengduo Zou, Yiren Zhao, YuKai Qin, and Lidong Yu recurring, and a collaboration score of 39. Output, impact, and activity axes are unscored because the underlying bibliometric data is insufficient; this profile is name-resolved from public bibliographic records.

Collaboration pattern

Collaboration is compact: 6 distinct coauthors appear across the two cataloged papers, with Jiaxing Chen, Hengduo Zou, Yiren Zhao, YuKai Qin, and Lidong Yu recurring. The radar's collaboration score is 39, on the lower end. Bolin Gao occupies the last author position on both 2026 preprints without a corresponding-author flag, a pattern consistent with a small, stable single-group team iterating on one end-to-end driving pipeline rather than a broad external network.

Research keywords

end-to-end autonomous drivingrisk-aware occupancyoccupancy predictiontrajectory planningbird's-eye view representationNAVSIM benchmarkPDMScollision ratenuScenesmultimodal trajectory prediction

Highlights

Risk-aware occupancy cuts GenAD collision rates by 35–53%
In 'Risk-Aware Occupancy for Safety-Oriented End-to-End Autonomous Driving', Evolve's review finds the 35–53% GenAD collision-rate drops promising for safety-oriented end-to-end driving, while flagging that the small-count open-loop metrics lack statistical testing and the evaluation builds on nuScenes with bird's-eye-view occupancy representations.
PDMS lifted 0.937→0.951 on NAVSIM via risk-aware occupancy
RRDrive adds risk-aware occupancy to a register-based NAVSIM planner, lifting PDMS from 0.937 to 0.951. Evolve's review credits the integration of risk reasoning into trajectory planning, but cautions that the headline gains rely on oracle future-derived risk labels, a dependency that limits real-world deployment claims.

Timeline

2026arXiv preprint 'Risk-Aware Occupancy for Safety-Oriented End-to-End Autonomous Driving' appears in the Evolve catalog, evaluating risk-aware occupancy on nuScenes with GenAD collision-rate drops of 35–53% and a medium reproducibility rating.
2026arXiv preprint 'Driving on Registers, Reasoning on Risk' adds risk-aware occupancy to a register-based NAVSIM planner, lifting PDMS from 0.937 to 0.951, also rated medium for reproducibility readiness.
2026Both cataloged 2026 papers list Bolin Gao in the last (senior) author position without a corresponding-author flag, consistent with a leading supervisory role in the group's end-to-end driving work at Tsinghua University.

Papers in the Evolve catalog

Frequent co-authors

Jiaxing Chen×2 Hengduo Zou×2 Yiren Zhao×2 Yiren Zhao×1 YuKai Qin×1 Lidong Yu×1

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.

Dimension scores are computed only from the bibliometric snapshot Evolve holds; output, impact, and activity axes have insufficient data and are deliberately unscored.
Both cataloged papers are 2026 arXiv preprints; no citation counts, venue acceptance, or peer-review status is available in this dataset.
No ORCID or country is recorded; this profile is name-resolved from public bibliographic data and the listed Tsinghua University affiliation.
The reproducibility score (55) rests on only 2 evaluated papers, both rated medium readiness, so it should be read as a thin estimate.