Share your thoughts, 1 month free Claude Pro on usSee more

Knowledge-Based Visual Question Answering on E-VQA Single-Hop

59.9Accuracy

PMSR

Updated 28d ago

Evaluation Results

Method	Links
PMSR 2025.08		59.9
OMGM 2025.08		51.2
OMGM 2025.08		50.2
DBAgent (SFT: E-VQA) 2026.04		46.5
PMSR 2025.08		46.4
DBAgent (SFT: Mixed) 2026.04		46
DBAgent (SFT: InfoSeek) 2026.04		45.1
PMSR 2025.08		40.9
CgRAG 2026.05		40.4
VLM-PRF 2026.04		40.1
Wiki-PRF(w/ RL) 2025.08		40.1
VLM-PRF w/ RL 2026.05		40.1
CgRAG 2026.05		39.9
CgRAG 2026.05		39.9
CgRAG 2026.05		39.8
MMKB-RAG 2025.08		39.7
MMKB-RAG 2026.05		39.7
CgRAG 2026.05		39.6
KU-RAG 2026.05		38.3
VLM-PRF 2026.04		37.1
ReflectiVA 2026.04		36.8
EchoSight 2025.08		35.5
ReflectiVA 2025.08		35.5
ReflectiVA 2026.05		35.5
DPR 2026.04		29.1
DPR_V+T 2026.05		29.1
VLM-PRF 2026.05		28.9
RET-2 2025.08		28.5
ReflectVA 2026.04		28
GPT-4V 2026.04		26.9
GPT-4V 2026.05		26.8
EchoSight 2026.05		26.4
DBAgent (w/o SFT) 2026.04		23.6
LLaVA-NeXT-8B 2026.05		22.5
LLaVA-NeXT-7B 2026.05		22.1
Wiki-LLaVA 2025.08		21.8
Qwen2.5-VL-7B (Base) 2026.04		21.7
Qwen2.5-VL-7B 2026.05		21.7
EchoSight 2026.04		19.4
EchoSight 2026.05		19.4
Wiki-LLaVA 2026.05		18.3
Qwen2.5-VL-3B 2026.05		17.9
Wiki-LLaVA 2026.04		17.7
Wiki-LLaVA 2026.05		17.7
LLaVA-v1.5 2026.04		16.3
LLaVA-1.5 2026.05		16.3
Qwen2-VL-7B 2026.05		16.2
LLaVA-1.5 2026.05		16
BLIP-2 2026.04		12.6
BLIP-2 2026.05		12.6
InstructBLIP 2026.04		11.9
InstructBLIP 2026.05		11.9