DreamerV3's imagined rollouts are insensitive to friction changes that cause real gait collapse, revealing that world models extrapolate kinematically rather than dynamically.
hub Baseline reference
Rethinking the Open-Loop Evaluation of End-to-End Autonomous Driving in nuScenes
Baseline reference. 57% of citing Pith papers use this work as a benchmark or comparison.
37
Pith papers citing it
9
external citations · Pith
Baseline
57% of classified citations
abstract
Modern autonomous driving systems are typically divided into three main tasks: perception, prediction, and planning. The planning task involves predicting the trajectory of the ego vehicle based on inputs from both internal intention and the external environment, and manipulating the vehicle accordingly. Most existing works evaluate their performance on the nuScenes dataset using the L2 error and collision rate between the predicted trajectories and the ground truth. In this paper, we reevaluate these existing evaluation metrics and explore whether they accurately measure the superiority of different methods. Specifically, we design an MLP-based method that takes raw sensor data (e.g., past trajectory, velocity, etc.) as input and directly outputs the future trajectory of the ego vehicle, without using any perception or prediction information such as camera images or LiDAR. Our simple method achieves similar end-to-end planning performance on the nuScenes dataset with other perception-based methods, reducing the average L2 error by about 20%. Meanwhile, the perception-based methods have an advantage in terms of collision rate. We further conduct in-depth analysis and provide new insights into the factors that are critical for the success of the planning task on nuScenes dataset. Our observation also indicates that we need to rethink the current open-loop evaluation scheme of end-to-end autonomous driving in nuScenes. Codes are available at https://github.com/E2E-AD/AD-MLP.
hub tools
citation-role summary
baseline 4
background 3
citation-polarity summary
representative citing papers
By adding future visual state prediction and a dedicated inverse kinematics diffusion network that uses only visual boundary conditions, a 0.5B driving VLA recovers visual grounding and matches 7-8B models on NAVSIM-v2 and nuScenes.
Bench2Drive-Robust is a new closed-loop benchmark that evaluates end-to-end autonomous driving models under deployment perturbations from camera failures, ego-state errors, and compute delays, showing substantial performance degradation beyond image-level tests.
The SNG framework and SNG-VLA model enable end-to-end driving systems to better incorporate global navigation for state-of-the-art route following without auxiliary perception losses.
ReCogDrive unifies VLM scene understanding with a diffusion planner reinforced by DiffGRPO to reach state-of-the-art results on NAVSIM and Bench2Drive benchmarks.
OmniSpace is a plug-and-play method that improves spatial reasoning in MLLMs for AV by injecting camera pose, using epipolar attention across views, and distilling 3D geometric knowledge to overcome weak cross-view correspondence and depth estimation.
PersonaDrive retrieves style-specific human driving demonstrations to condition a single VLA backbone for diverse closed-loop driving agents, reporting 4.6% and 2.5% driving score gains over baselines on Bench2Drive with style consistency within 2%.
VLGA introduces geometry as a fourth modality in VLA models via pointmap regression loss, reporting SOTA open-loop and closed-loop driving metrics on nuScenes and Bench2Drive.
Primary-path enclosure plus τ_exp-aligned fusion on a Dual-SoC AD-ECU yields 296 ms mean shutter-to-planner latency within a 350 ms budget while co-running modular and E2E paths.
VeriDrive introduces a verifiable counterfactual supervision framework using a Perception-Evaluation-Revision chain and validator-guided correction to generate cost-efficient structured data for vision-language driving models, showing metric gains on nuScenes.
Action Diffusion Transformer generates multimodal driving actions via diffusion and nearest-neighbor selection, claiming SOTA on Bench2Drive with 10x lower latency.
VECTOR-DRIVE uses shared self-attention with semantic-aware expert routing of tokens to VL and trajectory experts plus flow-matching action decoding to reach 88.91 driving score on Bench2Drive.
VLADriver-RAG reaches a new state-of-the-art Driving Score of 89.12 on Bench2Drive by retrieving structure-aware historical knowledge through spatiotemporal semantic graphs and Graph-DTW alignment.
HERMES++ unifies 3D scene understanding and future geometry prediction in driving scenes via BEV representations, LLM-enhanced queries, a temporal link, and joint geometric optimization.
ProDrive couples a query-centric planner with a BEV world model for end-to-end ego-environment co-evolution, enabling future-outcome assessment that improves safety and efficiency over reactive baselines on NAVSIM v1.
The primary OL-CL gap in end-to-end autonomous driving arises from objective mismatch creating structural inability to model reactive behaviors, which a test-time adaptation method can mitigate.
Orion-Lite uses latent feature distillation and trajectory supervision to create a vision-only model that surpasses its LLM-based teacher on closed-loop Bench2Drive evaluation, achieving a new SOTA driving score of 80.6.
Conditioning speed planning on the predicted path and relabeling synthetic cut-ins yields SOTA Bench2Drive scores (DS 89.07, SR 73.18%).
SpaceDrive replaces textual coordinate tokens with shared 3D positional encodings in a VLM driving planner, achieving state-of-the-art open-loop planning on nuScenes and 78.02 Driving Score on Bench2Drive.
AutoVLA unifies semantic reasoning and trajectory planning in one autoregressive VLA model for end-to-end autonomous driving by tokenizing trajectories into discrete actions and using GRPO reinforcement fine-tuning to adaptively reduce unnecessary reasoning.
LiloDriver uses LLMs and memory-augmented planning in a four-stage pipeline to outperform rule-based and learning-based methods on both common and rare scenarios in the nuPlan benchmark.
DriveMoE applies scene-specialized Vision MoE and skill-specialized Action MoE to a VLA baseline to achieve SOTA closed-loop performance on Bench2Drive.
ORION reports 77.74 Driving Score and 54.62% Success Rate on Bench2Drive, outperforming prior end-to-end methods by 14.28 DS and 19.61% SR through unified VQA and planning optimization.
EMMA is an end-to-end multimodal LLM that converts camera data into trajectories, objects, and road graphs via text prompts and reports state-of-the-art motion planning on nuScenes plus competitive detection results on Waymo.
citing papers explorer
-
Imagined Rollouts are Kinematic, Not Dynamic: A Diagnosis of Long-Horizon World-Model Failure
DreamerV3's imagined rollouts are insensitive to friction changes that cause real gait collapse, revealing that world models extrapolate kinematically rather than dynamically.
-
Grounding Driving VLA via Inverse Kinematics
By adding future visual state prediction and a dedicated inverse kinematics diffusion network that uses only visual boundary conditions, a 0.5B driving VLA recovers visual grounding and matches 7-8B models on NAVSIM-v2 and nuScenes.
-
Bench2Drive-Robust: Benchmarking Closed-Loop Autonomous Driving under Deployment Perturbations
Bench2Drive-Robust is a new closed-loop benchmark that evaluates end-to-end autonomous driving models under deployment perturbations from camera failures, ego-state errors, and compute delays, showing substantial performance degradation beyond image-level tests.
-
Unveiling the Surprising Efficacy of Navigation Understanding in End-to-End Autonomous Driving
The SNG framework and SNG-VLA model enable end-to-end driving systems to better incorporate global navigation for state-of-the-art route following without auxiliary perception losses.
-
ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving
ReCogDrive unifies VLM scene understanding with a diffusion planner reinforced by DiffGRPO to reach state-of-the-art results on NAVSIM and Bench2Drive benchmarks.
-
OmniSpace: Efficient Geometry Awareness for Autonomous Vehicles MLLMs
OmniSpace is a plug-and-play method that improves spatial reasoning in MLLMs for AV by injecting camera pose, using epipolar attention across views, and distilling 3D geometric knowledge to overcome weak cross-view correspondence and depth estimation.
-
PersonaDrive: Human-Style Retrieval-Augmented VLA Agents for Closed-Loop Driving Simulation
PersonaDrive retrieves style-specific human driving demonstrations to condition a single VLA backbone for diverse closed-loop driving agents, reporting 4.6% and 2.5% driving score gains over baselines on Bench2Drive with style consistency within 2%.
-
VLGA: Vision-Language-Geometry-Action Models for Autonomous Driving
VLGA introduces geometry as a fourth modality in VLA models via pointmap regression loss, reporting SOTA open-loop and closed-loop driving metrics on nuScenes and Bench2Drive.
-
An Exposure-Time-Aligned Primary-Path Architecture for Autonomous-Driving ECUs
Primary-path enclosure plus τ_exp-aligned fusion on a Dual-SoC AD-ECU yields 296 ms mean shutter-to-planner latency within a 350 ms budget while co-running modular and E2E paths.
-
VeriDrive: Verifiable Counterfactual Supervision for Cost-Efficient Vision-Language Planning
VeriDrive introduces a verifiable counterfactual supervision framework using a Perception-Evaluation-Revision chain and validator-guided correction to generate cost-efficient structured data for vision-language driving models, showing metric gains on nuScenes.
-
Multimodal Action Diffusion for Robust End-to-End Autonomous Driving
Action Diffusion Transformer generates multimodal driving actions via diffusion and nearest-neighbor selection, claiming SOTA on Bench2Drive with 10x lower latency.
-
VECTOR-Drive: Tightly Coupled Vision-Language and Trajectory Expert Routing for End-to-End Autonomous Driving
VECTOR-DRIVE uses shared self-attention with semantic-aware expert routing of tokens to VL and trajectory experts plus flow-matching action decoding to reach 88.91 driving score on Bench2Drive.
-
VLADriver-RAG: Retrieval-Augmented Vision-Language-Action Models for Autonomous Driving
VLADriver-RAG reaches a new state-of-the-art Driving Score of 89.12 on Bench2Drive by retrieving structure-aware historical knowledge through spatiotemporal semantic graphs and Graph-DTW alignment.
-
HERMES++: Toward a Unified Driving World Model for 3D Scene Understanding and Generation
HERMES++ unifies 3D scene understanding and future geometry prediction in driving scenes via BEV representations, LLM-enhanced queries, a temporal link, and joint geometric optimization.
-
ProDrive: Proactive Planning for Autonomous Driving via Ego-Environment Co-Evolution
ProDrive couples a query-centric planner with a BEV world model for end-to-end ego-environment co-evolution, enabling future-outcome assessment that improves safety and efficiency over reactive baselines on NAVSIM v1.
-
BridgeSim: Unveiling the OL-CL Gap in End-to-End Autonomous Driving
The primary OL-CL gap in end-to-end autonomous driving arises from objective mismatch creating structural inability to model reactive behaviors, which a test-time adaptation method can mitigate.
-
Orion-Lite: Distilling LLM Reasoning into Efficient Vision-Only Driving Models
Orion-Lite uses latent feature distillation and trajectory supervision to create a vision-only model that surpasses its LLM-based teacher on closed-loop Bench2Drive evaluation, achieving a new SOTA driving score of 80.6.
-
AlignDrive: Aligned Lateral-Longitudinal Planning for End-to-End Autonomous Driving
Conditioning speed planning on the predicted path and relabeling synthetic cut-ins yields SOTA Bench2Drive scores (DS 89.07, SR 73.18%).
-
SpaceDrive: Infusing Spatial Awareness into VLM-based Autonomous Driving
SpaceDrive replaces textual coordinate tokens with shared 3D positional encodings in a VLM driving planner, achieving state-of-the-art open-loop planning on nuScenes and 78.02 Driving Score on Bench2Drive.
-
AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning
AutoVLA unifies semantic reasoning and trajectory planning in one autoregressive VLA model for end-to-end autonomous driving by tokenizing trajectories into discrete actions and using GRPO reinforcement fine-tuning to adaptively reduce unnecessary reasoning.
-
LiloDriver: A Lifelong Learning Framework for Closed-loop Motion Planning in Long-tail Autonomous Driving Scenarios
LiloDriver uses LLMs and memory-augmented planning in a four-stage pipeline to outperform rule-based and learning-based methods on both common and rare scenarios in the nuPlan benchmark.
-
DriveMoE: Mixture-of-Experts for Vision-Language-Action Model in End-to-End Autonomous Driving
DriveMoE applies scene-specialized Vision MoE and skill-specialized Action MoE to a VLA baseline to achieve SOTA closed-loop performance on Bench2Drive.
-
ORION: A Holistic End-to-End Autonomous Driving Framework by Vision-Language Instructed Action Generation
ORION reports 77.74 Driving Score and 54.62% Success Rate on Bench2Drive, outperforming prior end-to-end methods by 14.28 DS and 19.61% SR through unified VQA and planning optimization.
-
EMMA: End-to-End Multimodal Model for Autonomous Driving
EMMA is an end-to-end multimodal LLM that converts camera data into trajectories, objects, and road graphs via text prompts and reports state-of-the-art motion planning on nuScenes plus competitive detection results on Waymo.
-
DriveVLM: The Convergence of Autonomous Driving and Large Vision-Language Models
DriveVLM adds vision-language models with scene description, analysis, and hierarchical planning modules to autonomous driving, paired with a hybrid DriveVLM-Dual system tested on nuScenes and SUP-AD datasets and deployed on a production vehicle.
-
What Probing Reveals about Autonomous Driving: Linking Internal Prediction Errors to Ego Planning
Autonomous driving policies with strong closed-loop performance frequently lack timely internal predictions of surrounding vehicles during near-collision events, and causal correction of prediction errors leads to improved ego planning.
-
DriveReward: A Comprehensive Dataset and Generative Vision-Language Reward Model for Autonomous Driving
Creates DriveReward dataset with counterfactual annotations and a 1B VLM reward model that outperforms larger VLMs on driving tasks and matches rule-based rewards in RL and trajectory scoring.
-
SparseWorld: Enhancing End-to-End Autonomous Driving via World Models with Sparse Scene Representation
SparseWorld is a sparse world model with a Sparse Dreamer module that performs autoregressive rollout of future instances to refine motion prediction and planning, reporting 0.05% collision rate on nuScenes open-loop metrics.
-
HEAT: Heterogeneous End-to-End Autonomous Driving via Trajectory-Guided World Models
HEAT uses a trajectory-driven learning paradigm and a world model predicting future latent features from ego actions to enable a single unified end-to-end autonomous driving model to perform well across heterogeneous domains on nuScenes, NAVSIM, and Waymo benchmarks.
-
DriveSafer: End-to-End Autonomous Driving with Safety Guidance
DriveSafer reduces catastrophic failures (PDMS=0) by 48% and drivable-area compliance failures by over 65% versus DiffusionDrive on the NAVSIM benchmark by combining training-time safety constraints with inference-time guidance.
-
Can Attribution Predict Risk? From Multi-View Attribution to Planning Risk Signals in End-to-End Autonomous Driving
Attribution statistics derived from multi-view inputs in end-to-end planners can predict planning risks, with reported Spearman correlation of 0.30 with trajectory error and AUROC of 0.77 for collision detection.
-
OVPD: A Virtual-Physical Fusion Testing Dataset of OnSite Auton-omous Driving Challenge
OVPD is a new virtual-physical fusion dataset with 20 testing clips totaling nearly 3 hours of multi-modal autonomous driving data for closed-loop evaluation.
-
Artificial Intelligence for Modeling and Simulation of Mixed Automated and Human Traffic
This survey synthesizes AI techniques for mixed autonomy traffic simulation and introduces a taxonomy spanning agent-level behavior models, environment-level methods, and cognitive/physics-informed approaches.
-
DriveStack-VLA: Render-Teacher Alignment for BEV-Based DeepStack Vision-Language-Action Model
DriveStack-VLA injects BEV into VLM decoder, aligns real and rasterized image focus, and adds head-based trajectory self-critique, reporting 91.6 PDMS on NAVSIMv1 and 79.49 driving score on Bench2Drive.
-
DeepSight: Long-Horizon World Modeling via Latent States Prediction for End-to-End Autonomous Driving
DeepSight uses parallel latent feature prediction in BEV for long-horizon world modeling and adaptive text reasoning to reach state-of-the-art closed-loop performance on the Bench2drive benchmark.
-
Conditional Flow-VAE for Safety-Critical Traffic Scenario Generation
A conditional flow matching model generates realistic safety-critical traffic scenarios by turning nominal scenes into dangerous rollouts using combined simulation and real data.
-
DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving
A hybrid imitation-plus-reinforcement diffusion planner generates more diverse multi-mode trajectories for end-to-end autonomous driving, with a new diversity metric used for evaluation.