List your company

Visual Document Question Answering on LongDocURL (Filtered)

67.79AccuracyOracle
43.485249.795156.10562.4149Sep 2, 2025

Evaluation results

MethodLinks
OracleTop-k retrieval=3, Input modality=I + T, Judge model=Qwen2.5-7B-Instruct, Backbone Vision Language Model=Qwen2.5-VL-7B-Instruct2025.0967.79
OracleTop-k retrieval=3, Input modality=I, Judge model=Qwen2.5-7B-Instruct, Backbone Vision Language Model=Qwen2.5-VL-7B-Instruct2025.0964.81
CMRAGTop-k retrieval=3, Judge model=Qwen2.5-7B-Instruct, Backbone Vision Language Model=Qwen2.5-VL-7B-Instruct2025.0948.18
SigLIPTop-k retrieval=3, Judge model=Qwen2.5-7B-Instruct, Backbone Vision Language Model=Qwen2.5-VL-7B-Instruct2025.0947.66
CLIP-L/14-336Top-k retrieval=3, Judge model=Qwen2.5-7B-Instruct, Backbone Vision Language Model=Qwen2.5-VL-7B-Instruct2025.0944.42