Research map / Probabilistic and scientific modeling / Molecules and proteins
GuideFlow: Constraint-Guided Flow Matching for Planning in End-to-End Autonomous Driving
Lin Liu, Caiyan Jia, Guanyi Yu, Ziying Song, Junqiao Li, Feiyang Jia, Peiliang Wu, Xiaoshuai Hao, Yadan Luo
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 plan anchor conditioning signal (ablation) | GuideFlow baseline · NavSim Navhard (no scorer) | EPDMS: 23.1 -> 29.0 | Improved |
| use goal point or driving command instead of plan anchor (ablation) | GuideFlow · NavSim Navhard (no scorer) | EPDMS: 29.0 -> 28.6 (goal point), 27.1 (command) | Hurt |
| add Constraining the Velocity Field (CVF) (ablation) | GuideFlow baseline · NavSim Navhard (no scorer) | EPDMS: 23.1 -> 24.5 | Improved |
| add Constraining the Flow States (CF) (ablation) | GuideFlow baseline · NavSim Navhard (no scorer) | EPDMS: 23.1 -> 25.1 | Improved |
| add Refining the Flow by EBM (RFE) (ablation) | GuideFlow baseline · NavSim Navhard Stage2 (OOD) | Stage2 EPDMS: 40.0 -> 45.3 | Improved |
| combine CF and RFE (ablation) | GuideFlow baseline · NavSim Navhard (no scorer) | EPDMS: 23.1 -> 27.1 | Improved |
| add reward-as-style conditioning with EP=1 (ablation) | GuideFlow (CF+RFE) · NavSim Navhard (no scorer) | EPDMS: 27.1 -> 26.3 | Hurt |
| add reward-as-style conditioning with EP=1 (ablation) | GuideFlow (CF+RFE) · NavSim Navhard (no scorer) | Ego Progress: 79.6 -> 82.3 | Improved |
| increase velocity-correction weight lambda from 0.1 to 0.5 (ablation) | CVF · NavSim Navhard | EPDMS: 24.5 -> 13.5 | Hurt |
| reduce sampling steps K from 100 to 10 (ablation) | GuideFlow · NavSim Navhard | EPDMS: 27.1 -> 21.9 | Hurt |
| start flow-state correction at kc=40 vs 10 (ablation) | CF · NavSim Navhard | EPDMS: 24.2 (kc=10) -> 27.1 (kc=40) -> 25.0 (kc=50) | Mixed |
| use GuideFlow instead of SparseDrive on out-of-domain adversarial scenes | SparseDrive baseline · ADV-NuScenes | Avg collision rate (%): 1.02 -> 0.73 | Improved |
Later papers that compared with this method
1 later accepted paper reports GuideFlow as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- DriveFuture: Future-Aware Latent World Models for Autonomous Driving (NeurIPS 2026 · better on 1 of 1 shared result)
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
- GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving (CVPR 2025 · similarity 0.89)
- Flow Matching-Based Autonomous Driving Planning with Advanced Interactive Behavior Modeling (NeurIPS 2025 · similarity 0.87)
- ChainFlow-VLA: Causal Flow Planning with Vision-Language Models (NeurIPS 2026 · similarity 0.86)
- On the Road to Portability: Compressing End-to-End Motion Planner for Autonomous Driving (CVPR 2024 · similarity 0.86)
- Model-Based Policy Adaptation for Closed-Loop End-to-end Autonomous Driving (NeurIPS 2025 · similarity 0.85)
- Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving (NeurIPS 2024 · similarity 0.85)
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.
