end-to-end-driving

NVIDIA Alpamayo 1 Nano is an open 10B reasoning VLA model for autonomous vehicles that pairs driving trajectories with Chain-of-Causation reasoning.

AlpaSim is an open-source autonomous vehicle simulation platform designed for development and testing of end-to-end AV policies

[ICCV'23] Hidden Biases of End-to-End Driving Models & A starter kit for the CARLA leaderboard 2.0.

Repo of "GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving"

NVIDIA Alpamayo 1.5 Nano is an open 10B reasoning VLA model for autonomous vehicles with reinforcement-learning enhanced reasoning, navigation guidance, and visual question answering.

NeuroNCAP benchmark for end-to-end autonomous driving

NVIDIA Alpamayo 2 Super is an open 34B multi-task foundation model designed to supercharge autonomous vehicle development.

[CVPR26] LEAD: Minimizing Learner–Expert Asymmetry in End-to-End Driving

[ECCV 2024] Embodied Understanding of Driving Scenarios

AlpaGym is a reinforcement-learning framework for end-to-end autonomous-driving policies.

🏆 Official implementation of LangCoop: Collaborative Driving with Natural Language

A curated list of papers on post-training for end-to-end autonomous driving: distillation, preference alignment, reinforcement learning, and test-time refinement.

Using diffusion model to reach controllable end-to-end driving with Carla simulation environment.

AutoMoE: a PyTorch Mixture‑of‑Experts self‑driving stack for CARLA with trained perception experts, a gating network, and a trajectory policy, plus datasets and training/inference scripts.

VLA ≠ VLM. Side-by-side viewer running NVIDIA Alpamayo R1 (vision-language-action) alongside Qwen2.5-VL (vision-language) on the same 44-sec SF dashcam clip at 5 Hz. 220 paired traces. Surfaces what an action-trained model sees that a scene-trained model doesn't, and vice versa.

Closed-loop CARLA benchmark for authority-aware autonomous driving: can a model obey a human traffic director (police officer, construction flagger, ambulance) when the human directive overrides the traffic rule? Tracks A/B/C, oracle-calibrated MARSHAL-Graded score.

Converts Waymo Open Dataset end to end driving data to h5 format

𝒮𝒟-2 · System Deviation Diagnosis — a robustness diagnosis framework for end-to-end (E2E) autonomous driving. Decomposes the driving pipeline (vision → semantic → planning → control → outcome), measures stage-wise deviation between clean and stress CARLA runs, and localizes where robustness first collapses (InterFuser, TransFuser).

E2E model divergence analysis for autonomous driving