arXiv:2609.22762v1 Announce Type: new Abstract: Generative world-action models (WAMs) jointly generate future video and vehicle actions, while their action branches remain primarily optimized by expe...
The paper introduces CoLT-Drive, a 3,536-sample counterfactual long‑tail benchmark for evaluating decision‑level driving affordance prediction, which tests whether models can infer how rare objects affect an ego vehicle’s high‑level actions. It also proposes KPA, a knowledge‑preserving adaptation framework that combines structured prompting, expert merging, and a regime‑aware LoRA mixture‑of‑experts module to improve small VLMs on driving tasks. Experiments show KPA achieves 60.8% pair accuracy on CoLT‑Drive, outperforming the Qwen3‑VL‑2B baseline and LoRA SFT while keeping competitive in‑domain performance.
arXiv:2606. 31844v1 Announce Type: cross Abstract: A local-to-global context mismatch arises when autoregressive traffic simulators trained on ego-centric driving logs are deployed in globally observable closed-loop environments.
The paper presents an end‑to‑end system that converts driving footage into dynamic vision sensor (DVS) event streams, augments training with simulated DVS data, and trains a convolutional spiking neural network (Conv‑SNN) to classify pedestrian crossing intent as crossing or non‑crossing. The Conv‑SNN, trained with a class‑balanced loss and surrogate‑gradient learning, achieves high accuracy and F1 scores on JAAD and CARLA DVS datasets, outperforming or matching prior frame‑based methods while operating on sparse temporal representations. The study details architectural choices, neuron dynamics, and training protocols, and provides a convergence analysis and domain‑transfer evaluation.
OPTED is a method for on‑policy fine‑tuning of end‑to‑end driving models that separates reinforcement learning from the policy update. A privileged teacher trained with RL on vectorized inputs (HD‑maps and bounding boxes) supervises the pre‑trained student during closed‑loop post‑training. Applied to the camera‑based models TransFuser and VaVAM in AlpaSim, OPTED boosts driving scores by 1.6× and 9.5×, respectively, while requiring roughly three orders of magnitude fewer simulator interactions than direct RL post‑training.
The paper studies how Bird's‑Eye‑View (BEV) maps predicted by Cross‑View Transformers (CVT) can be used directly as inputs to a Behavior‑Cloning (BC) driving policy in the CARLA simulator. It introduces a six‑channel BEV representation and a Kernel Density Estimation (KDE) weighting scheme to focus learning on underrepresented maneuvers. Closed‑loop tests show that the KDE‑weighted model is the only predicted‑BEV agent to finish an episode without infractions, highlighting that global segmentation scores are poor proxies for driving performance and that prediction quality at critical geometries, especially the route channel, is key to reliable navigation.
RiskTraf introduces a risk-extrapolated residual learning approach for multi-variate traffic flow prediction, leveraging raw flow, speed, and occupancy data from the new PEMSB-3V benchmark. The method freezes a trained spatio-temporal backbone and adds a lightweight residual head that learns from historical speed and occupancy to correct flow predictions across different traffic regimes. Experiments show consistent improvements over various backbones and outperform existing debiasing and distribution-shift adaptation techniques.
arXiv:2608.28656v1 Announce Type: cross Abstract: Behavior-cloned Vision-Language-Action (VLA) driving policies struggle with rare rule-governed maneuvers at signalized intersections. Braking and lau...
The paper introduces Endpoint-Constrained Optimization (ECO), a lightweight postprocessing layer that corrects intermediate waypoints of end-to-end driving policies while preserving the predicted endpoint. ECO does not require maps, privileged simulator state, or additional training, and can be applied to a wide range of waypoint-emitting policies. Experiments on two closed-loop simulators show that ECO significantly improves closed-loop performance, achieving top results in the HUGSIM Closed-Loop Driving Challenge and boosting scene scores on AlpaSim.
MILER is an end‑to‑end reinforcement learning framework that achieves zero‑shot sim‑to‑real transfer for autonomous driving in unstructured environments. It uses a custom semantic mid‑level representation (MLR) simulator for offline training, and during deployment it processes real camera and LiDAR data with BEVFusion to produce a compatible bird’s‑eye‑view representation. The policy’s actions are applied via a trajectory‑alignment strategy, allowing the system to drive 17.3 km on a 3.0 km test track without human intervention, all running on a Jetson AGX Orin.
arXiv:2606. 03159v1 Announce Type: cross Abstract: As autonomous vehicle capabilities advance, the safe evaluation of driving policies in long-tail scenarios remains a critical bottleneck.
arXiv:2607. 20549v1 Announce Type: new Abstract: Trajectory datasets used in ADAS evaluation are heavily biased toward routine driving; genuine vehicle-to-vehicle conflict events are rare, and the rarer the event, the higher the cost when an ADAS system fails to handle it.