Foundation Models for V2X-based Cooperative Autonomous Driving

Introduction

The 7th edition of the full-day DriveX workshop explores advances in Foundation Models and 3D Perception in Cooperative Autonomous Driving (CAD). This workshop brings together leading researchers and practitioners to discuss cutting-edge developments in large language models (LLMs), vision-language models (VLMs), vision-language action models (VLAs), and their applications to autonomous driving systems. Topics include 3D object detection, semantic segmentation, sensor fusion, V2X communication, and cooperative perception.

We explore methods to enhance scene understanding, perception accuracy, dataset curation, and novelty detection. By uniting experts across perception, V2X, and foundation model domains, this workshop aims to foster innovation in cooperative autonomous driving and intelligent transportation systems. The workshop addresses critical challenges in multi-modal sensor data fusion, vehicle-infrastructure coordination, and intelligent transportation systems that leverage both onboard and roadside sensing capabilities.

This year, we expanded our focus with the addition of V2X applications, exploring real-world vehicle-to-infrastructure connectivity that extends past collaborative perception. The workshop provides a platform for discussing V2X for localization, tolling, road safety, monitoring, and data analytics, bridging the gap between theoretical advances and practical deployment in intelligent transportation systems. Through keynote presentations, panel discussions, paper presentations, and challenge tracks, DriveX 2026 creates a comprehensive forum for advancing the state-of-the-art in foundation model-driven cooperative autonomous driving.

Topics of Interest

3D Environment Perception

• 3D Scene Understanding
• 3D Instance Segmentation
• 3D Occupancy Prediction
• 3D Detection and Tracking

Cooperative Perception

• V2X Communication
• Vehicle-Infrastructure Fusion
• Roadside & ITS Sensors (RSUs)
• Multi-modal Sensor Data Fusion

Foundation Models

• LLM-assisted Perception & Prediction
• Vision-Language Models (VLMs)
• FMs for Dataset Curation & Labeling
• FMs for Accident & Novelty Detection

Schedule (Tentative)

Final schedule, room allocation, and speaker order will be announced closer to the workshop date.

Confirmed Keynote Speakers

• V2I-Based Cooperative Perception
  Infrastructure–vehicle fusion using TUMTraf-V2X. Focus on cooperative 3D detection and tracking with infrastructure-mounted LiDAR, radar, and cameras, emphasizing occlusion handling, long-range awareness, and reliability under real-world conditions.

• Natural Language Instruction for Human Interaction and Vision-Language Navigation
  Built upon doScenes. Participants design models for natural language instructions and visual language navigation to facilitate research on human-vehicle instruction interactions.

• Multi-Agent Reasoning
  Using MDrive teams explore a cooperative driving benchmark for end-to-end closed-loop multi-agent systems.

Competition Timeline

• Competition Announcement: Apr 13, 2026
• Competition Start: Apr 15, 2026 (10 weeks)
• Submission Deadline: July, 1 (23:59 PST)
• Notification to Participants: July 7, 2026

Top-performing teams will be invited to present at the workshop and will receive money prizes ($100) and award certificates. Detailed rules, baselines, and submission instructions are available on the official challenge page.

Organizers

Invited Program Committee

Sponsors

DriveX 2026 welcomes sponsorship from industry, startups, and institutions interested in foundation models, cooperative perception, simulation, and large-scale autonomous driving systems.

For sponsorship opportunities, please contact: wz@ucla.edu.