Call for participation: ImageCLEF 2025 Multimodal Reasong Lab
We’re excited to announce the MultimodalReasoning shared task as part of CLEF 2025, focusing on the reasoning capabilities of Vision-Language Models (VLMs).
Motivation: While VLMs excel at basic tasks like image captioning and simple visual question answering, they struggle with complex reasoning. This task challenges participants to explore and improve their models' ability to reason through multilingual, multimodal inputs.
Task Description: Participants will address Multilingual Visual Question Answering (VQA) by identifying the correct answer from multiple options, given an image.
Evaluation Metric: Accuracy will determine the model's performance.
Why Participate?
Advance the field of multimodal AI reasoning.
Benchmark your systems on challenging multilingual data.
Share your insights at CLEF 2025 in Madrid, Spain!
Key Dates:
Registration Opens: 20 Dec 2024
Registration Closes: 25 Apr 2025
Test Data Release: 24 Mar 2025
Submission Deadline: 10 May 2025
We invite you to register and join this exciting challenge!
📍 More Info & Registration Details: https://www.imageclef.org/2025/multimodalreasoning
Let’s shape the future of reasoning in AI together!
Best regards, Task Organizing Team, MultimodalReasoning