Research map / Vision-language and multimodal / Robot action and navigation
LMGenDrive: Bridging Multimodal Understanding and Generative World Modeling for End-to-End Driving
Hao Shao, Letian Wang, Yang Zhou, Yuxuan Hu, Zhuofan Zong, Steven Waslander, Wei Zhan, Hongsheng LI
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add world generator with world queries (ablation) | LMGenDrive · LangAuto | DS: 53.4 -> 62.2 | Improved |
| add learnable action queries instead of autoregressive action prediction (ablation) | LMGenDrive · LangAuto | DS: 58.7 -> 62.2 | Improved |
| add stage-1 driving visual pre-training (ablation) | LMGenDrive · LangAuto | DS: 54.9 -> 62.2 | Improved |
| add stage-3 multi-step long-horizon training (ablation) | LMGenDrive · LangAuto | DS: 55.6 -> 62.2 | Improved |
| add multi-view fusion to world generator (ablation) | LMGenDrive · World generation | FVD: 371 -> 286 | Improved |
| use 64 world queries instead of 32 (ablation) | LMGenDrive · World generation | FID: 10.1 -> 6.3 | Improved |
| use 64 world queries instead of 16 (ablation) | LMGenDrive · World generation | FVD: 424 -> 286 | Improved |
| add stage-2 single-step training before stage 3 (fixed budget) (ablation) | LMGenDrive · LangAuto-Short | DS: 67.9 -> 77.1 | Improved |
| add stage-3 training with fixed iteration budget (ablation) | LMGenDrive · LangAuto-Short | DS: 72.3 -> 77.1 | Improved |
| add stage-3 training with fixed iteration budget (ablation) | LMGenDrive · LangAuto-Tiny | DS: 80.1 -> 84.1 | Improved |
| lengthen autoregressive video horizon from 8 to 128 frames | LMGenDrive world generator · World generation | FVD: 286 -> 981 | Hurt |
| use LMGenDrive instead of BEVDriver | LMGenDrive · LangAuto | DS: 48.9 -> 62.2 | Improved |
Closest papers by meaning
- Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (CVPR 2024 · similarity 0.90)
- DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generation (AAAI 2025 · similarity 0.89)
- Generating Multimodal Driving Scenes via Next-Scene Prediction (CVPR 2025 · similarity 0.89)
- FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving (NeurIPS 2025 · similarity 0.89)
- DriveGPT4-V2: Harnessing Large Language Model Capabilities for Enhanced Closed-Loop Autonomous Driving (CVPR 2025 · similarity 0.89)
- SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Driving (CVPR 2026 · similarity 0.89)
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.
