Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Video Question Answering on VCGPT (test)
Loading...
44.35
Model-as-Judge Score
M3KG-RAG
38.3492
39.9071
41.465
43.0229
Dec 23, 2025
Model-as-Judge Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Model-as-Judge Score
M3KG-RAG
MLLM=Qwen2.5-Omni
2025.12
44.35
VAT-KG
MLLM=Qwen2.5-Omni
2025.12
43.5
VTKG
MLLM=Qwen2.5-Omni
2025.12
42.96
M2ConceptBase
MLLM=Qwen2.5-Omni
2025.12
42.78
None
MLLM=Qwen2.5-Omni
2025.12
42.21
Wikidata
MLLM=Qwen2.5-Omni
2025.12
40.82
M3KG-RAG
MLLM=VideoLLaMA2
2025.12
39.92
VAT-KG
MLLM=VideoLLaMA2
2025.12
39.42
M2ConceptBase
MLLM=VideoLLaMA2
2025.12
39.31
None
MLLM=VideoLLaMA2
2025.12
39.09
VTKG
MLLM=VideoLLaMA2
2025.12
38.88
Wikidata
MLLM=VideoLLaMA2
2025.12
38.58
Feedback
Search any
task
Search any
task
