Pith. sign in

REVIEW 27 cited by

DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2411.15139 v3 pith:3VHLC7UJ submitted 2024-11-22 cs.CV cs.RO

DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving

classification cs.CV cs.RO
keywords diffusiondiffusiondrivedrivingmodelpolicydenoisingmulti-modesteps
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Recently, the diffusion model has emerged as a powerful generative technique for robotic policy learning, capable of modeling multi-mode action distributions. Leveraging its capability for end-to-end autonomous driving is a promising direction. However, the numerous denoising steps in the robotic diffusion policy and the more dynamic, open-world nature of traffic scenes pose substantial challenges for generating diverse driving actions at a real-time speed. To address these challenges, we propose a novel truncated diffusion policy that incorporates prior multi-mode anchors and truncates the diffusion schedule, enabling the model to learn denoising from anchored Gaussian distribution to the multi-mode driving action distribution. Additionally, we design an efficient cascade diffusion decoder for enhanced interaction with conditional scene context. The proposed model, DiffusionDrive, demonstrates 10$\times$ reduction in denoising steps compared to vanilla diffusion policy, delivering superior diversity and quality in just 2 steps. On the planning-oriented NAVSIM dataset, with the aligned ResNet-34 backbone, DiffusionDrive achieves 88.1 PDMS without bells and whistles, setting a new record, while running at a real-time speed of 45 FPS on an NVIDIA 4090. Qualitative results on challenging scenarios further confirm that DiffusionDrive can robustly generate diverse plausible driving actions. Code and model will be available at https://github.com/hustvl/DiffusionDrive.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 27 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Decaying Turbulence and the Riemann Hypothesis: The number theory behind the infinite-time singularity

    hep-th 2026-04 unverdicted novelty 8.0

    Freely decaying incompressible turbulence possesses a universal Euler-ensemble attractor whose continuum Mellin spectrum is controlled by the non-trivial zeros of the Riemann zeta function, producing an infinite-time ...

  2. 4DLidarOpen: An Open 4D FMCW Lidar Dataset for Motion-Aware Autonomous Driving

    cs.RO 2026-05 unverdicted novelty 7.0

    4DLidarOpen is a new open dataset providing synchronized 4D FMCW Lidar velocity measurements, multi-Lidar and camera data, and 3D bounding-box annotations with track IDs to support benchmarks on 3D detection, BEV segm...

  3. Unveiling the Surprising Efficacy of Navigation Understanding in End-to-End Autonomous Driving

    cs.RO 2026-04 unverdicted novelty 7.0

    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.

  4. Multi-Modal Manipulation via Multi-Modal Policy Consensus

    cs.RO 2025-09 unverdicted novelty 7.0

    A policy that factorizes into modality-specific diffusion models combined by a learned router network for adaptive multi-modal robotic manipulation.

  5. ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving

    cs.CV 2025-06 unverdicted novelty 7.0

    ReCogDrive unifies VLM scene understanding with a diffusion planner reinforced by DiffGRPO to reach state-of-the-art results on NAVSIM and Bench2Drive benchmarks.

  6. AlphaDrive: Unleashing the Power of VLMs in Autonomous Driving via Reinforcement Learning and Reasoning

    cs.CV 2025-03 unverdicted novelty 7.0

    AlphaDrive uses GRPO-based RL rewards and two-stage SFT+RL training on VLMs to improve autonomous driving planning performance and efficiency while producing emergent multimodal capabilities.

  7. D$^3$-MoE:Dual Disentangled Diffusion Mixture-of-Experts for Style-Controllable End-to-End Autonomous Driving

    cs.RO 2026-06 unverdicted novelty 6.0

    D³-MoE disentangles style and physical axes with diffusion and self-supervised MoE experts to produce style-controllable trajectories, reporting SOTA 88.2 PDMS on NAVSIM.

  8. OneDrive: Unified Multi-Paradigm Driving with Vision-Language-Action Models

    cs.CV 2026-04 unverdicted novelty 6.0

    OneDrive unifies heterogeneous decoding in a single VLM transformer decoder for end-to-end driving, achieving 0.28 L2 error and 0.18 collision rate on nuScenes plus 86.8 PDMS on NAVSIM.

  9. TaCarla: A comprehensive benchmarking dataset for end-to-end autonomous driving

    cs.RO 2026-02 conditional novelty 6.0

    TaCarla releases 2.85M CARLA Leaderboard 2.0 frames with nuScenes-style sensors, multi-task annotations, planning baselines, and a text-based rarity score.

  10. From Representational Complementarity to Dual Systems: Synergizing VLM and Vision-Only Backbones for End-to-End Driving

    cs.RO 2026-02 conditional novelty 6.0

    A VLM-based and a vision-only end-to-end planner are behaviorally complementary in a long tail of driving scenarios; selecting the better trajectory lifts NAVSIM PDMS from 90.80 to 92.10 at modest compute.

  11. Self-Imitated Diffusion Policy for Efficient and Robust Visual Navigation

    cs.RO 2026-01 conditional novelty 6.0

    SIDP trains a diffusion policy for visual navigation by reward-weighting its own sampled trajectories, improving success rate and cutting inference latency.

  12. AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning

    cs.CV 2025-06 unverdicted novelty 6.0

    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...

  13. Using Ensemble Diffusion to Estimate Uncertainty for End-to-End Autonomous Driving

    cs.RO 2025-05 unverdicted novelty 6.0

    EnDfuser replaces point-estimate trajectory planning with ensemble diffusion in a single attention-pooling transformer module to model posterior trajectory uncertainty and improve safety in end-to-end autonomous driving.

  14. Uni-NaVid: A Video-based Vision-Language-Action Model for Unifying Embodied Navigation Tasks

    cs.RO 2024-12 unverdicted novelty 6.0

    Uni-NaVid unifies diverse embodied navigation tasks into one video-based vision-language-action model trained on 3.6 million samples from four sub-tasks, achieving state-of-the-art performance on benchmarks and real-w...

  15. ASSCG: Just-Right Gating over Chattering for Fast-Slow LLM Planning in Autonomous Driving

    cs.RO 2026-06 unverdicted novelty 5.0

    ASSCG is an RWKV-based adaptive gate trained with SFT and GRPO-style RL that makes Query/Cache/Drop decisions for slow LLM guidance in fast-slow autonomous driving planners, improving scores and cutting latency on nuP...

  16. Diffusion Forcing Planner: History-Annealed Planning with Time-Dependent Guidance for Autonomous Driving

    cs.RO 2026-06 unverdicted novelty 5.0

    Diffusion Forcing Planner applies heterogeneous joint diffusion with time-dependent noise and classifier-free guidance on history segments to generate stable, controllable motion plans for autonomous driving on nuPlan.

  17. HEAT: Heterogeneous End-to-End Autonomous Driving via Trajectory-Guided World Models

    cs.RO 2026-05 unverdicted novelty 5.0

    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 ...

  18. Latency Analysis and Optimization of Alpamayo 1 via Efficient Trajectory Generation

    cs.AI 2026-05 unverdicted novelty 5.0

    Redesigning Alpamayo 1 to single-reasoning and optimizing diffusion action generation cuts inference latency by 69.23% while preserving trajectory diversity and prediction quality.

  19. SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Model

    cs.CV 2026-04 unverdicted novelty 5.0

    SpanVLA reduces action generation latency via flow-matching conditioned on history and improves robustness by training on negative-recovery samples with GRPO and a dedicated reasoning dataset.

  20. LADY: Linear Attention for Autonomous Driving Efficiency without Transformers

    cs.AI 2025-12 conditional novelty 5.0

    LADY shows that an end-to-end driving model using only linear attention can match transformer-based planners on NAVSIM/Bench2Drive while fusing arbitrary-length historical sensor frames at constant per-frame cost.

  21. DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving

    cs.CV 2025-07 unverdicted novelty 5.0

    DIVER uses RL-guided diffusion to produce diverse feasible trajectories from one ground-truth path, addressing mode collapse in imitation learning for autonomous driving.

  22. FocalAD: Local Motion Planning for End-to-End Autonomous Driving

    cs.AI 2025-06 conditional novelty 5.0

    FocalAD adds an ego-local graph interactor and focal loss to prioritize decision-critical neighbors, yielding lower collision rates than prior methods on nuScenes, Bench2Drive, and especially the Adv-nuScenes robustness set.

  23. Enhancing Text-to-Image Diffusion Transformer via Split-Text Conditioning

    cs.CV 2025-05 unverdicted novelty 5.0

    DiT-ST converts complete-text captions into split-text primitives via LLMs and injects them hierarchically across denoising stages to reduce semantic confusion in DiT-based text-to-image generation.

  24. Intend, Reflect, Refine: An Adaptive Multimodal Reflection Framework for Autonomous Driving

    cs.CV 2026-06 unverdicted novelty 4.0

    IRR-Drive adds an adaptive multimodal reflection step (text intention plus predicted future BEV) that lets a VLA model self-correct its trajectory plan according to scene complexity and reports SOTA on NAVSIM.

  25. CLEAR: Cognition and Latent Evaluation for Adaptive Routing in End-to-End Autonomous Driving

    cs.RO 2026-06 unverdicted novelty 4.0

    CLEAR achieves state-of-the-art PDMS of 93.7 on NAVSIM v1 by combining single-step VAE latent drift with Qwen 3.5-guided adaptive scheduling and trajectory scoring for end-to-end driving.

  26. DeepSight: Long-Horizon World Modeling via Latent States Prediction for End-to-End Autonomous Driving

    cs.CV 2026-05 unverdicted novelty 4.0

    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.

  27. Decaying Turbulence and the Riemann Hypothesis: The number theory behind the infinite-time singularity

    hep-th 2026-04 unverdicted novelty 3.0

    Decaying turbulence converges to an Euler ensemble attractor whose statistics are governed by an arithmetic structure from the Farey sequence of coprime fractions.