AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

RAVENEA: A Benchmark for Multimodal Retrieval-Augmented Visual Culture Understanding

Jiaang Li, Yifei Yuan, Wenyan Li, Mohammad Aliannejadi, Daniel Hershcovich, Anders Søgaard, Ivan Vulić, Wenxuan Zhang, Paul Pu Liang, Yang Deng, Serge Belongie

ICLR 2026 · Vision-language and multimodal › Visual question answering · approach: Multimodal LLM question answering · 6 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fine-tune CLIP with culture-aware contrastive loss on RAVENEACLIP retriever · RAVENEA retrievalP@1: 60.87 -> 72.05Improved
fine-tune SigLIP with culture-aware contrastive lossSigLIP retriever · RAVENEA retrievalMRR: 68.62 -> 70.95Improved
fine-tune generative models (VisualBERT, VL-T5, LLaVA-OV-7B) as retrievers instead of contrastive modelsretriever · RAVENEA retrievalMRR: 75.44 -> 61.33Hurt
add Ravenea-CLIP retrieval augmentation to VLMs17 VLMs · cVQAaccuracy: 67.1 -> 73.3Improved
add Ravenea-CLIP retrieval augmentation to VLMs17 VLMs · cICRegionScore: 57.5 -> 68.8Improved
use culture-aware fine-tuned CLIP instead of frozen CLIP for retrievalVLMs with RAG · cVQAaccuracy: 71.9 -> 73.3Improved
use culture-aware fine-tuned CLIP instead of frozen CLIP for retrievalVLMs with RAG · cICRegionScore: 65.1 -> 68.8Improved
add Ravenea-CLIP retrieval to a small VLMQwen3-VL-2B · cVQAaccuracy: +31.6%Improved
add Ravenea-CLIP retrieval to a large VLMQwen3-VL-32B · cICRegionScore: no gainNo clear effect
add culturally relevant retrieved documents instead of unrelated documents (ablation)Ravenea-CLIP · cVQA (12 VLMs)accuracy: 60.6 -> 71.2Improved
add culturally relevant retrieved documents instead of random characters (ablation)Ravenea-CLIP · cVQA (12 VLMs)accuracy: 62.3 -> 71.2Improved
add culturally relevant retrieved documents instead of unrelated documents (ablation)Ravenea-CLIP · cIC (12 VLMs)RegionScore: 48.8 -> 67.7Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy