Call for papers @ GRAIL-V, CVPR 2026

[Jyotika Singh]

Dear AI researchers and practitioners,

Computer vision is moving beyond passive perception. The next frontier is Agentic Intelligence -- systems that don't just "see" data, but actively reason, plan, and operate over it to solve complex workflows.

We invite you to shape this future by submitting your work to the GRAIL-V (Grounded Retrieval & Agentic Intelligence for Vision-Language) workshop at CVPR 2026 in Denver.

This is not just another workshop. GRAIL-V is the dedicated venue for bridging the gap between research capability and deployment reliability. We are looking for work that addresses the "hard parts" of AI agents: reliability, grounding, and real-world execution. We aim to bring CV, NLP, IR and HCI communities together for the next break-through in Agentic Vision eco-system.

Key Themes & Topics:

• Agentic Workflows: Planning, tool routing, and multi-step reasoning.
• Multimodal Retrieval: Search across images, video, UI, and documents.
• Grounding & Evidence: Citation provenance and audit-ready faithfulness.
• Generative Action: Editing and manipulating content (Image/Video/Text) as part of a workflow.
• Benchmarks & Evaluation: Reproducible methods for measuring success.
• Real-World Constraints: Latency, cost-efficiency, and production benchmarks.

Speakers:

• Kristen Grauman (UT Austin)
• Mohit Bansal (UNC, Chapel Hill)
• Dan Roth (UPenn)
• Scott Yih (FAIR, Meta)
• Sujith Ravi (Oracle AI)

Submission Snapshot:

• 📅 Submission Deadline: March 5, 2026
• 📢 Author Notification: March 18, 2026
• 📍 Conference Dates: June 3–7, 2026 (Denver, USA)
• 📝 Submission Link: https://openreview.net/group?id=thecvf.com/CVPR/2026/Workshop/GRAIL-V
• 🌐 Workshop Website: https://grailworkshops.github.io/

Note: Accepted papers will be included in the CVPR Proceedings.

We look forward to seeing your latest breakthroughs in Denver.

Best regards,

The GRAIL-V Organizing Team 

(Oracle AI, Arizona State University, University of Utah, KAIST, Adaption Labs)