InfoSeek

InfoSeek is a benchmark dataset used to evaluate knowledge-based visual question answering (KB-VQA) systems by providing relevant external documents for conditioning answer generation.

Papers using InfoSeek (17)

ProMSA:Progressive Multimodal Search Agents for Knowledge-Based Visual Question Answering2026Ground Then Rank: Revisiting Knowledge-Based VQA with Training-Free Entity Identification2026Enhancing Visual Question Answering with Multimodal LLMs via Chain-of-Question Guided Retrieval-Augmented Generation2026Region-R1: Reinforcing Query-Side Region Cropping for Multi-Modal Re-Ranking2026WikiSeeker: Rethinking the Role of Vision-Language Models in Knowledge-Based Visual Question Answering2026Learning to Search: A Decision-Based Agent for Knowledge-Based Visual Question Answering2026Wiki-R1: Incentivizing Multimodal Reasoning for Knowledge-based VQA via Data and Sampling Curriculum2026When RAG Hurts: Diagnosing and Mitigating Attention Distraction in Retrieval-Augmented LVLMs2026MaS-VQA: A Mask-and-Select Framework for Knowledge-Based Visual Question Answering2026CC-VQA: Conflict- and Correlation-Aware Method for Mitigating Knowledge Conflict in Knowledge-Based Visual Question Answering2026ReAG: Reasoning-Augmented Generation for Knowledge-based Visual Question Answering2025Knowledge-based Visual Question Answer with Multimodal Processing, Retrieval and Filtering2025Progressive Multimodal Search and Reasoning for Knowledge-Intensive Visual Question Answering2025OMGM: Orchestrate Multiple Granularities and Modalities for Efficient Multimodal Retrieval2025Cross-modal Retrieval for Knowledge-based Visual Question Answering2024 · 10 citesCan Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions?2023 · 2 citesEchoSight: Advancing Visual-Language Models with Wiki Knowledge2024