List your company

Knowledge-based Visual Question Answering on Infoseek (test)

53.4AccuracyProMSA
10.86421.90732.9543.993Dec 2, 2024Mar 7, 2025Jun 10, 2025Sep 13, 2025Dec 17, 2025Mar 22, 2026Jun 26, 2026

Evaluation results

MethodLinks
ProMSARetriever=BGE+EVA-CLIP, Model=Qwen3-VL-8B, Category=Retrieval-Augmented Models2026.0653.453.653.3
ProMSARetriever=BGE+EVA-CLIP, Model=Qwen2.5-VL-7B, Category=Retrieval-Augmented Models2026.0649.248.849.6
CC-VQARetriever=EVA-CLIP-8B, Model=Qwen2.5-VL-7B, Category=Retrieval-Augmented Models2026.0645.144.746.1
REALRetriever=EVA-CLIP-8B, Model=Qwen3-VL-8B, Category=Retrieval-Augmented Models2026.0644.143.145.1
MaS-VQARetriever=EVA-CLIP-8B, Model=Qwen3-VL-8B, Category=Retrieval-Augmented Models2026.0643.843.743.9
ProMSARetriever=BGE+EVA-CLIP, Model=Qwen3-VL-2B, Category=Retrieval-Augmented Models2026.0643.644.143
DeepEyesV2*Retriever=BGE+EVA-CLIP, Model=Qwen2.5-VL-7B, Category=Search-Agent Models2026.0641.341.640.8
MMSearch-R1*Retriever=BGE+EVA-CLIP, Model=Qwen2.5-VL-7B, Category=Search-Agent Models2026.0639.740.339.3
EchoSight†Retriever=EVA-CLIP-8B, Model=Qwen3-VL-8B, Category=Retrieval-Augmented Models2026.0632.432.632.2
RA-VQAv2 w/ PreFLMR2024.1230.7——
EchoSightRetriever=EVA-CLIP-8B, Model=Mistral-7B, Category=Retrieval-Augmented Models2026.0630.43030.7
LamRATraining=Simultaneous retrieval and VQA using LoRA2024.1228.8——
Qwen3-VL-8BRetriever=None, Model=None, Category=Zero-shot MLLMs2026.0625.725.925.5
Qwen2.5-VL-7BRetriever=None, Model=None, Category=Zero-shot MLLMs2026.0623.722.824.1
Qwen3-VL-2BRetriever=None, Model=None, Category=Zero-shot MLLMs2026.0618.918.619.2
GPT-4VRetriever=None, Model=None, Category=Zero-shot MLLMs2026.0614.61514.3
BLIP-2Retriever=None, Model=Flan-T5, Category=Zero-shot MLLMs2026.0612.512.712.3