FAU at ImageCLEF 2026 Task on Multimodal Reasoning Robust Candidate Scoring and Concise Multilingual Visual Answering
2026-08-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionMachine Learning
AI summaryⓘ
The authors built a system for answering visual questions about scientific and educational images that have text, charts, and formulas. Instead of letting the model write answers freely, they made it score fixed answer choices and used smart tricks like combining multiple guesses and cleaning up output. For open-ended questions, they enhanced images and carefully controlled the text generation to produce neat, correct answers. Their system did very well in a competition without extra training, showing that clever ways of handling model output are key to reliability.
Visual Question AnsweringImageCLEFVision-Language ModelsMultimodal ReasoningScore FusionDeterministic DecodingPromptingEnsemblingPost-processingEvaluation Metrics
Authors
Mohamed Basem, Vincent Christlein
Abstract
We present our ImageCLEF 2026 Multimodal Reasoning system for the Visual Multiple Choice Question Answering (Visual MCQ) and Visual Open Question Answering (Visual OpenQA) subtasks. The challenge requires reliable reasoning over multilingual educational and scientific images with dense text, diagrams, charts, tables, formulas, and units, while enforcing strict answer formats. Our central finding is that robust output control is as important as model choice. For Visual MCQ, we replace fragile free-form generation with direct candidate label scoring from vision-language model logits, then combine complementary runs through score fusion and voting. For Visual OpenQA, we use image enhancement, concise final answer prompting, deterministic decoding, and targeted post-processing to remove reasoning traces and formatting artifacts. Without task-specific model training, our official submissions achieved third place in Visual MCQ with 0.7108 accuracy and first place in Visual OpenQA with 0.6488 COMET, 0.1391 BLEU, 0.2762 ROUGE L, and 0.2383 METEOR. The results highlight the practical value of inference engineering: careful scoring, ensembling, prompting, and cleanup can turn strong VLMs into reliable competition systems.