AI Paper Atlas

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

MOJITO: Modal Joint Learning for Unified End-to-End Autonomous Driving

Zhijing Cheng, Xuancheng Zhang, Donglin Di, Lei Fan, Baorui Ma, Hao Li, Xun Yang

ECCV 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 4 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
add LiDAR input with PillarGroup tokenization to the camera-only model (ablation)MOJITO · NAVSIM-v1 navtestPDMS: 86.8 -> 88.9Improved
use PillarGroup instead of FPS+KNN LiDAR tokenization (ablation)MOJITO · NAVSIM-v1 navtestPDMS: 86.1 -> 88.9Improved
add LiDAR with FPS+KNN tokenization to camera-only model (ablation)MOJITO · NAVSIM-v1 navtestPDMS: 86.8 -> 86.1Hurt
use joint self-attention instead of cross-attention (sensor tokens attend to action tokens) (ablation)MOJITO · NAVSIM-v1 navtestPDMS: 85.7 -> 88.9Improved
increase blocks from 3 to 12 (ablation)MOJITO · NAVSIM-v1 navtestPDMS: 80.5 -> 88.9Improved
increase blocks from 3 to 6 (ablation)MOJITO · NAVSIM-v1 navtestPDMS: 80.5 -> 80.5No clear effect
increase blocks from 6 to 9 (ablation)MOJITO · NAVSIM-v1 navtestPDMS: 80.5 -> 83.0Improved
increase blocks from 9 to 12 (ablation)MOJITO · NAVSIM-v1 navtestPDMS: 83.0 -> 88.9Improved
add RL fine-tuning to ReCogDrive-BaseReCogDrive-Base · NAVSIM-v1 navtestPDMS: 86.5 -> 90.8Improved
add RL fine-tuning to AutoVLAAutoVLA · NAVSIM-v1 navtestPDMS: 80.5 -> 89.1Improved
add RL fine-tuning to AdaThinkDriveAdaThinkDrive · NAVSIM-v1 navtestPDMS: 86.2 -> 90.3Improved
remove LiDAR branch (camera-only) on navhardMOJITO · NAVSIM-v2 navhardEPDMSHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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