AI & ML
Instructions Before Actions: Prospective and Retrospective Language Supervision for Driving Vision-Language-Action Data
arXiv cs.RO0 views
arXiv:2611.20631v1 Announce Type: cross Abstract: Driving vision-language-action (VLA) models can accept navigation instructions, yet predicting a driver's trajectory does not establish that a model follows the instruction. We study how driving-language supervision is collected and evaluated, distinguishing prospective instructions issued before an action from retrospective labels written after observing it. We develop ADVLAT-Engine to synchronize spoken instructions, road-scene video, and vehicle trajectories, and compare three navigation apps on 48 replayed routes. On 87 maneuvers announced by both Google Maps and a passenger, the passenger uses more scene-dependent language. We evaluate four open driving VLAs before and after fine-tuning on prospective navigation-app drives. Fine-tuning reduces passenger-drive trajectory error by 23-59%, but instruction effects remain inconsistent. At real intersections, counterfactual instructions reveal stronger responsiveness: fine-tuned Alpamayo~2 Super gains 31 and 18 accuracy points at 15 and 5~m, while Qwen-Drive follows alternative maneuvers more reliably with a discrete command than with a sentence. We additionally collected retrospective annotations from four people for the driving recordings and describe controlled comparisons of annotation content, timing, and downstream learning; those analyses are pending. Our findings separate maneuver prediction from instruction-following and establish a framework for comparing prospective and retrospective supervision.
Source:https://arxiv.org/abs/2611.20631PreviousVoxSAOR: Semantic-Aware Occupancy Robustification via Aleatoric Uncertainty ModelingNextSTA: A Split Temporal Adversary Method for Evaluating Robustness of Multimodal RL-based Driving Policies
Related posts

Claude went down twice on 13 November: what broke, what kept working, and what to do next time
0dev.to

Five entry types that belong in your AI agent's memory file
0dev.to
PS5 Pro vs Base PS5: Performance, Upgrades, and Is It Worth It?
0dev.to
Black high school students sue after AI mistakes Doritos bag for gun
0Hacker News
Thống kê đối đầu Toulouse vs Lens: Sự chênh lệch rõ nét về hiệu suất chiến thuật
0dev.to Data
Thống kê đối đầu Auxerre vs Monaco: Ưu thế vượt trội kéo dài hơn một thập kỷ của đội bóng Công quốc
0dev.to Data