Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long-context Document Understanding on LongDocURL
Loading...
64.5
Accuracy
GPT-4o
2.932
18.916
34.9
50.884
Apr 15, 2026
Accuracy
Updated 3d ago
Evaluation Results
Method
Method
Links
Accuracy
GPT-4o
Backbone=-, Param=-, P...
2026.04
64.5
Doc-V* (GRPO)
Backbone=Qwen2.5-VL, P...
2026.04
56.3
Doc-V* (SFT)
Backbone=Qwen2.5-VL, P...
2026.04
53
URaG
Backbone=Qwen2.5-VL, P...
2026.04
52.2
MoLoRAG
Backbone=Qwen2.5-VL, P...
2026.04
51.9
Gemini-1.5-Pro
Backbone=-, Param=-, P...
2026.04
50.9
VRAG-RL
Backbone=Qwen2.5-VL, P...
2026.04
44.9
VisRAG
Backbone=MiniCPM-V 2.6...
2026.04
41.9
VDocRAG
Backbone=Phi3-Vision,...
2026.04
39.8
InternVL3
Backbone=InternViT / Q...
2026.04
38.7
Qwen2.5-VL (RAG Top-5)
Backbone=Qwen2.5-VL, P...
2026.04
37.8
M3DocRAG
Backbone=Qwen2-VL, Par...
2026.04
35.1
Qwen2.5-VL (Baseline)
Backbone=Qwen2.5-VL, P...
2026.04
32.9
mPLUG-DocOwl2
Backbone=ViT / LLaMa,...
2026.04
5.3
Feedback
Search any
task
Search any
task