Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety & Grounding on M3TAVR (test)
Loading...
8.1
Hallucination Rate
R-CGA + InternVL3.5-8B
6.908
14.954
23
31.046
Jun 25, 2026
Hallucination Rate
mIoU
Updated 29d ago
Evaluation Results
Method
Method
Links
Hallucination Rate
mIoU
R-CGA + InternVL3.5-8B
Framework=TAVR-VLM, Ba...
2026.06
8.1
62.4
R-CGA + Qwen3-VL-8B
Framework=TAVR-VLM, Ba...
2026.06
8.7
60.3
Gemini-3 Pro
Paradigm=Closed-Source...
2026.06
11.5
-
Claude-4 Opus
Paradigm=Closed-Source...
2026.06
12.9
-
GPT-4o
Paradigm=Closed-Source...
2026.06
13.8
-
InternVL3.5-8B
Paradigm=Open-Source M...
2026.06
14.7
51.6
Qwen3-VL-8B
Paradigm=Open-Source M...
2026.06
17.9
50.1
LLaVA-Med-1.5
Paradigm=Open-Source M...
2026.06
24.2
45.4
METransformer
Paradigm=Report Genera...
2026.06
33.8
23.4
R2Gen-Mamba
Paradigm=Report Genera...
2026.06
34.6
21
MCLN-Transformer
Paradigm=Report Genera...
2026.06
37.9
17.5
Feedback
Search any
task
Search any
task