SV-WAM：面向端到端自动驾驶的高效环视世界-动作模型
Original title: SV-WAM: An Efficient Surround-View World-Action Model for End-to-End Autonomous Driving
ResearchAI 62
Source: arXiv cs.ROPublish time unverified
arXiv:2609.03602v1 Announce Type: cross Abstract: World models (WMs) have demonstrated strong potential for end-to-end autonomous driving by learning predictive representations of future scene dynamics. However, generating future videos during inference introduces substantial computational overhead, leading many recent driving WMs to adopt a single front camera as input for efficient deployment.