arXiv AI
Aligning Multi-Trajectory Supervision with Policy Optimization for VLA Driving
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Machine Learning
Jul 31
S-GRPO: Unified Post-Training for Large Vision-Language Models
arXiv:2604. 16557v2 Announce Type: replace Abstract: Current post-training methodologies for adapting Large Vision-Language Models (LVLMs) generally fall into two paradigms: Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL).
arXiv Computer Vision
Oct 12
SimVTA: Simulation-Grounded Video-Trajectory Alignment for Driving World Action Models
SimVTA is a post‑training framework that enhances driving world action models by aligning simulated future videos with candidate trajectories. It samples trajectories, renders high‑reward futures in a real‑world simulator using 3D Gaussian Splatting, and prioritizes video supervision for high‑error futures. The approach improves planning performance, raising PDMS from 86.2 to 91.3 on NAVSIM v1 and achieving strong results on NAVSIM v2 and HUGSIM.
arXiv AI
Oct 12
OGPO: Offline Goal-conditioned Policy Optimization for Recoverable Vision-Language-Action Models
OGPO is an Offline Goal‑conditioned Policy Optimization framework designed to improve Vision‑Language‑Action (VLA) models for robotic manipulation. It replaces traditional step‑wise action imitation with adaptive relabeling of trajectories using semantic key states and goal‑conditioned rewards, enabling policies to reach task‑progress states even from off‑demonstration states. Experiments on LIBERO, MetaWorld, LIBERO‑Plus, and the proposed LIBERO‑ReAct benchmark show that OGPO consistently outperforms standard supervised fine‑tuning in terms of robustness and recoverability under distribution shifts.
arXiv AI
Jun 24
Beyond Trajectory Imitation: Strategy-Guided Policy Optimization for LLM Reasoning
arXiv:2606. 24064v1 Announce Type: new Abstract: Distilling reasoning capabilities from strong to weak language models typically involves imitating specific solution trajectories, effectively transferring what to answer rather than how to reason.
arXiv Computer Vision
Oct 12
FPD-Drive: Future Policy Distillation for End-to-End Autonomous Driving
FPD-Drive is an end‑to‑end autonomous driving planner that uses future‑privileged multi‑target distillation to transfer diverse, high‑quality trajectory targets from a teacher model with access to future observations to a student policy. It also introduces safety‑aware score correction to improve candidate selection by adjusting rankings between safe and unsafe trajectories. The approach achieves state‑of‑the‑art performance on NAVSIM v1 and v2 benchmarks and demonstrates strong zero‑shot closed‑loop results on HUGSIM.
arXiv Machine Learning
Jun 11
OGPO: Sample Efficient Full-Finetuning of Generative Control Policies
arXiv:2605. 03065v2 Announce Type: replace Abstract: Generative control policies (GCPs), such as diffusion- and flow-based control policies, have emerged as effective parameterizations for robot learning.