• XPENG-PKU Research Breakthrough: In cooperation with Peking University, XPENG has developed FastDriveVLA—a novel visual token pruning framework that allows autonomous driving AI to “drive like a human” by concentrating only on essential information, resulting in a 7.5 – fold reduction in computational load.
• Top – Tier AI Recognition: The research has been accepted by AAAI 2026, one of the world’s leading AI conferences, which had a highly selective acceptance rate of only 17.6% this year.
• Accelerating L4 Autonomy: This achievement highlights XPENG’s full – stack capabilities in AI – driven mobility and promotes the industry towards the efficient and scalable deployment of next – generation autonomous driving systems.

GUANGZHOU, China, Dec. 28, 2025 — XPENG, in cooperation with Peking University, has had its paper “FastDriveVLA: Efficient End – to – End Driving via Plug – and – Play Reconstruction – based Token Pruning” accepted by AAAI 2026, one of the world’s top conferences in artificial intelligence. AAAI 2026 received 23,680 submissions, with only 4,167 papers accepted, an acceptance rate of just 17.6%.

The paper presents FastDriveVLA, an efficient visual token pruning framework specifically designed for end – to – end autonomous driving Vision – Language – Action (VLA) models. This work provides a new way for visual token pruning by enabling AI to “drive like a human”, focusing only on essential visual information and filtering out irrelevant data.

As AI large models evolve quickly, VLA models are being widely used in end – to – end autonomous driving systems because of their strong abilities in complex scene understanding and action reasoning. These models encode images into a large number of visual tokens, which serve as the basis for the model to “see” the world and make driving decisions. However, processing a large number of tokens increases the computational load on the vehicle, affecting inference speed and real – time performance.

Although visual token pruning has been recognized as a feasible method to speed up VLA inference, existing methods, whether based on text – visual attention or token similarity, have shown limitations in driving scenarios. To solve this problem, XPENG and PKU developed FastDriveVLA, a novel reconstruction – based token pruning framework inspired by how human drivers focus on relevant foreground information while ignoring non – critical background areas.

The method introduces an adversarial foreground – background reconstruction strategy that improves the model’s ability to identify and keep valuable tokens. On the nuScenes autonomous driving benchmark, FastDriveVLA achieved state – of – the – art performance across various pruning ratios. When the number of visual tokens was reduced from 3,249 to 812, the framework achieved nearly a 7.5 – fold reduction in computational load while maintaining high planning accuracy.

This is the second time this year that XPENG has been recognized at a top – tier global AI conference. In June, XPENG was the only Chinese automaker invited to speak at CVPR WAD, where it shared progress in autonomous driving foundation models. At its AI Day in November, XPENG unveiled the VLA 2.0 architecture, which removes the “language translation” step and enables direct Visual – to – Action generation, a breakthrough that redefines the traditional V – L – A pipeline.

These achievements reflect XPENG’s full – stack in – house capabilities, from model architecture design and training to distillation and vehicle deployment. Looking forward, XPENG remains dedicated to achieving L4 level autonomous driving to accelerate the integration of physical AI systems into vehicles, aiming to provide safe, efficient, and comfortable intelligent driving experiences to users around the world.

About XPENG
XPENG is committed to leading the transformation of future mobility through technological exploration, positioning itself as “Explorer of Future Mobility”. Headquartered in Guangzhou, China, the company operates R & D centers in Beijing, Shanghai, Shenzhen, Zhaoqing, and Yangzhou, and has established intelligent manufacturing bases in Zhaoqing and Guangzhou.

XPENG pursues a global strategy for research, development, and sales, with an R & D center in the United States and subsidiaries in multiple European countries. The company adheres to full – stack in – house development of intelligent driver – assistance software and the development of core hardware, providing an excellent intelligent driving and riding experience for users.

On August 27, 2020, XPENG officially listed on the New York Stock Exchange (NYSE: XPEV), raising funds in an IPO that set a record at the time for the global new energy vehicle industry. On July 7, 2021, the company listed on the Hong Kong Stock Exchange (HKEX: 9868), becoming the first Chinese new – energy automaker to achieve dual primary listings in both Hong Kong and New York.

For more information, please visit .

Contacts:
For Media Enquiries: Alison Liang, XPENG PR Department
Email:  

SOURCE XPENG