A new VLA model called SI uses a four-step chain-of-thought to derive driving intent and applies it via classifier-free guidance to a flow-matching trajectory generator, showing competitive Waymo scores and intent-controllable plans.
hub
Poutine: Vision-language-trajectory pre-training and reinforcement learning post-training enable robust end-to-end autonomous driving
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
KITScenes LongTail supplies multimodal driving data and multilingual expert reasoning traces to benchmark models on rare scenarios beyond basic safety metrics.
DriveTeach-VLA adds Driving-aware Vision Distillation pretraining and 2D Trajectory-Guided Prompts to VLA models, then reports state-of-the-art results on NAVSIM and nuScenes.
Self-play DAgger training in a batched pixel renderer produces end-to-end driving policies that reach competitive performance on HUGSIM and NAVSIM-v2 after real-world adaptation and improve with more self-play compute.
NTR adds a self-distillation masked latent reconstruction objective that uses only scene tokens to reconstruct masked patch features, improving visual representation quality and planning performance in end-to-end autonomous driving.
ReasonBreak demonstrates up to 89% attack success on reasoning and 72% on trajectories in NVIDIA Alpamayo VLA models via black-box textual perturbations, introducing a reasoning-aware evaluation framework and benchmark for autonomous driving.
Fast-dDrive is a block-diffusion VLA that reports SOTA accuracy on WOD-E2E and nuScenes driving benchmarks together with 12x throughput over autoregressive baselines via section scaffolds and test-time averaging.
MindVLA-U1 is the first unified streaming VLA architecture that surpasses human drivers on WOD-E2E planning metrics while matching VA latency and preserving language interfaces.
DRIV-EX generates fluent counterfactual scene descriptions by using gradient-optimized embeddings only as a guide for controlled text decoding, producing more reliable explanations than baselines on transcribed highD driving data.
Alpamayo-R1 introduces a VLA model with a Chain of Causation dataset and multi-stage SFT-plus-RL training that reports 12% better planning accuracy and 35% fewer close encounters versus trajectory-only baselines in driving tasks.
CoPhy is a new RL framework that distills VLM cognition into BEV encoders, adds an auto-regressive BEV world model for action-conditioned future prediction, and optimizes policies via GRPO with dual physical-cognitive rewards, claiming SOTA on NAVSIM v1/v2.
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.
citing papers explorer
-
Action Emergence from Streaming Intent
A new VLA model called SI uses a four-step chain-of-thought to derive driving intent and applies it via classifier-free guidance to a flow-matching trajectory generator, showing competitive Waymo scores and intent-controllable plans.
-
LongTail Driving Scenarios with Reasoning Traces: The KITScenes LongTail Dataset
KITScenes LongTail supplies multimodal driving data and multilingual expert reasoning traces to benchmark models on rare scenarios beyond basic safety metrics.
-
Teaching Vision-Language-Action Models What to See and Where to Look
DriveTeach-VLA adds Driving-aware Vision Distillation pretraining and 2D Trajectory-Guided Prompts to VLA models, then reports state-of-the-art results on NAVSIM and nuScenes.
-
Scaling Self-Play for End-to-End Driving
Self-play DAgger training in a batched pixel renderer produces end-to-end driving policies that reach competitive performance on HUGSIM and NAVSIM-v2 after real-world adaptation and improve with more self-play compute.
-
NTR: Neural Token Reconstruction for Scene Token Bottleneck in End-to-End Driving
NTR adds a self-distillation masked latent reconstruction objective that uses only scene tokens to reconstruct masked patch features, improving visual representation quality and planning performance in end-to-end autonomous driving.
-
ReasonBreak: Probing Vulnerabilities in Reasoning-Enabled Vision-Language-Action Models for Autonomous Driving
ReasonBreak demonstrates up to 89% attack success on reasoning and 72% on trajectories in NVIDIA Alpamayo VLA models via black-box textual perturbations, introducing a reasoning-aware evaluation framework and benchmark for autonomous driving.
-
Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving
Fast-dDrive is a block-diffusion VLA that reports SOTA accuracy on WOD-E2E and nuScenes driving benchmarks together with 12x throughput over autoregressive baselines via section scaffolds and test-time averaging.
-
MindVLA-U1: VLA Beats VA with Unified Streaming Architecture for Autonomous Driving
MindVLA-U1 is the first unified streaming VLA architecture that surpasses human drivers on WOD-E2E planning metrics while matching VA latency and preserving language interfaces.
-
DRIV-EX: Counterfactual Explanations for Driving LLMs
DRIV-EX generates fluent counterfactual scene descriptions by using gradient-optimized embeddings only as a guide for controlled text decoding, producing more reliable explanations than baselines on transcribed highD driving data.
-
Alpamayo-R1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail
Alpamayo-R1 introduces a VLA model with a Chain of Causation dataset and multi-stage SFT-plus-RL training that reports 12% better planning accuracy and 35% fewer close encounters versus trajectory-only baselines in driving tasks.
-
Distill to Think, Foresee to Act: Cognitive-Physical Reinforcement Learning for Autonomous Driving
CoPhy is a new RL framework that distills VLM cognition into BEV encoders, adds an auto-regressive BEV world model for action-conditioned future prediction, and optimizes policies via GRPO with dual physical-cognitive rewards, claiming SOTA on NAVSIM v1/v2.
-
SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Model
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.