Guardado en:
Detalles Bibliográficos
Autores principales: Zhu, Ruiyang, He, Yuehan, Zheng, Boyuan, Zhao, Zesen, Chalhoub, Ahmad, Zhang, Qingzhao, Mao, Z. Morley
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2605.17284
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918507796496384
author Zhu, Ruiyang
He, Yuehan
Zheng, Boyuan
Zhao, Zesen
Chalhoub, Ahmad
Zhang, Qingzhao
Mao, Z. Morley
author_facet Zhu, Ruiyang
He, Yuehan
Zheng, Boyuan
Zhao, Zesen
Chalhoub, Ahmad
Zhang, Qingzhao
Mao, Z. Morley
contents End-to-end autonomous driving systems powered by Vision-Language-Action (VLA) models achieve strong performance on common driving scenarios, yet remain brittle in rare but safety-critical long-tail situations such as active construction zones and complex yielding geometries. In this paper, we present a method that addresses the long-tail challenging scenes beyond data scaling and model training. We introduce CLAP (Contrastive Latent-space Prompt optimization), a location-aware adaptation framework that augments a frozen VLA driving model with per-roadblock soft prompts, optimized from crowdsourced data and retrieved on demand via Vehicle-to-Everything (V2X) communication. Our approach rests on two observations from VLAs' latent space: (i) at the VLA's hidden-state layer, scenarios from the same roadblock cluster tightly and occupy compact regions of the latent space; and (ii) within a single roadblock, long-tail and normal frames are heavily intermixed in the latent representation, making it difficult to improve one without disturbing the other. CLAP addresses this via a two-stage pipeline: supervised contrastive learning to discover a roadblock-specific hard-scene direction, followed by directionally regularized prompt optimization that selectively improves challenging frames while preserving normal frame performance. On the NAVSIM benchmark with various state-of-the-art VLA backbones, CLAP reduces challenging scenario planning error by 24% with no regression on normal frames, significantly improving planning performance.
format Preprint
id arxiv_https___arxiv_org_abs_2605_17284
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle CLAP: Contrastive Latent-space Prompt Optimization for End-to-end Autonomous Driving
Zhu, Ruiyang
He, Yuehan
Zheng, Boyuan
Zhao, Zesen
Chalhoub, Ahmad
Zhang, Qingzhao
Mao, Z. Morley
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Robotics
End-to-end autonomous driving systems powered by Vision-Language-Action (VLA) models achieve strong performance on common driving scenarios, yet remain brittle in rare but safety-critical long-tail situations such as active construction zones and complex yielding geometries. In this paper, we present a method that addresses the long-tail challenging scenes beyond data scaling and model training. We introduce CLAP (Contrastive Latent-space Prompt optimization), a location-aware adaptation framework that augments a frozen VLA driving model with per-roadblock soft prompts, optimized from crowdsourced data and retrieved on demand via Vehicle-to-Everything (V2X) communication. Our approach rests on two observations from VLAs' latent space: (i) at the VLA's hidden-state layer, scenarios from the same roadblock cluster tightly and occupy compact regions of the latent space; and (ii) within a single roadblock, long-tail and normal frames are heavily intermixed in the latent representation, making it difficult to improve one without disturbing the other. CLAP addresses this via a two-stage pipeline: supervised contrastive learning to discover a roadblock-specific hard-scene direction, followed by directionally regularized prompt optimization that selectively improves challenging frames while preserving normal frame performance. On the NAVSIM benchmark with various state-of-the-art VLA backbones, CLAP reduces challenging scenario planning error by 24% with no regression on normal frames, significantly improving planning performance.
title CLAP: Contrastive Latent-space Prompt Optimization for End-to-end Autonomous Driving
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Robotics
url https://arxiv.org/abs/2605.17284