2026· Conference and Labs of the Evaluation Forum· pp. 3738-3745· 1 citation· 8 references
Computer Science
TL;DR
Analysis reveals that combining high-resolution memory optimization with a carefully designed QRA prompt template plays a crucial role in stimulating and enhancing the cross-lingual reasoning performance of small models, providing a new perspective for multimodal reasoning tasks.
Training-free adaptation of medical vision-language models (LVLMs) is a deployment requirement, yet visual in-context learning (ICL) injects modality- and style-mismatched demonstrations into medical contexts, teaching a frozen model the wrong answering style. We present MedICL, which curates demonstrations instead of...
Ji-jun Park, Soo-joon Choi, Seung-Min Han et al.· 0 citations
Many artwork questions begin with vision but cannot be answered by visual recognition alone. A model may need to connect an artwork to its creator, depicted subject, collection, or location, and then follow multiple facts to reach the answer. We introduce A RT H OP QA, a multimodal multi-hop question answering benchmar...
Xue-Feng Wei, Xuan Zhou, Y. Kato et al.· Volume 1· 0 citations
Interleaved multimodal in-context learning (ICL) lets frozen large vision-language models (LVLMs) acquire new tasks from a handful of image-text-answer demonstrations, yet it remains fragile: demonstrations bind to the wrong images, redundant exemplars crowd out informative ones, and noisy answers are copied verbatim i...
Y. Yılmaz, Elena Petrova, M. Kaya et al.· 0 citations
Thinking with images (TwI) empowers vision-language models (VLMs) to overcome perceptual bottlenecks by actively cropping and zooming into fine-grained visual details, promising both higher accuracy and interpretable, region-grounded reasoning. However, our analysis shows that the improved accuracy of TwI models does n...
Yu Zhao, Jia-Xin Zhang, Rohit Saxena et al.· 0 citations
Some questions about an image need commonsense or factual knowledge that its pixels do not contain. Large Language Models (LLMs) can supply this knowledge to Visual Question Answering (VQA) systems, but the depth at which their guidance should enter a vision-language model remains unclear. We prompt the 70-billion-para...
Aneerban Saha, Vijay Kumar, D. SameerMadhav et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.