AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Hidden Biases of End-to-End Driving Models

Bernhard Jaeger, Kashyap Chitta, Andreas Geiger

ICCV 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 101 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use target point conditioning instead of discrete navigation commands (ablation)TransFuser reproduction · CARLA validation routesRoute Completion: 56 -> 84Improved
add shift and rotation augmentation to NC-conditioned model (ablation)TransFuser with NC conditioning · CARLA validation routesDriving Score: 32 -> 35No clear effect
replace global average pooling with transformer decoder pooling (ablation)TransFuser reproduction · CARLA validation routesRoute Completion: 84 -> 93Improved
replace global average pooling with transformer decoder pooling (ablation)TransFuser reproduction · CARLA validation routesstatic collisions per km: 1.04 -> 0.55Improved
add shift and rotation augmentation to transformer decoder model (ablation)TransFuser reproduction · CARLA validation routesDriving Score: 43 -> 49Improved
predict path plus confidence-weighted speed classification instead of waypoints (ablation)TransFuser reproduction · CARLA validation routesDriving Score: 49 -> 50No clear effect
use argmax speed class instead of confidence-weighted speed (ablation)TransFuser reproduction with path output · CARLA validation routesDriving Score: 50 -> 40Hurt
add two-stage training with perception pre-training (ablation)TransFuser reproduction · CARLA validation routesDriving Score: 50 -> 54Improved
freeze pre-trained backbone in second training stage (ablation)TransFuser reproduction · CARLA validation routesDriving Score: 54 -> 44Hurt
triple dataset size from 185k to 555k frames (ablation)TransFuser reproduction · CARLA validation routesDriving Score: 54 -> 60Improved
double training epochs in a single stage (ablation)TransFuser reproduction · CARLA validation routesDriving ScoreNo clear effect
use TF++ instead of TransFuserTF++ · CARLA LAV validation townsDriving Score: 39 -> 70Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy