Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Active Question Answering on OpenEQA v1 (HM3D)
Loading...
85.1
LLM-Match
Human baseline
21.452
37.976
54.5
71.024
Dec 17, 2025
LLM-Match
LLM-Match SPL
Updated 4d ago
Evaluation Results
Method
Method
Links
LLM-Match
LLM-Match SPL
Human baseline
2025.12
85.1
-
R4
2025.12
74
63.37
3D-Mem
2025.12
52.6
42
GPT-4V
2025.12
41.8
7.5
GPT-4 w/ LLaVA-1.5
Vision-Language Model=...
2025.12
38.1
7
GPT-4
2025.12
35.5
-
GPT-4 w/ CG
Representation=Concept...
2025.12
34.4
6.5
GPT-4 w/ SVM
Representation=Sparse...
2025.12
34.2
6.4
LLaMA-2 w/ LLaVA-1.5
Vision-Language Model=...
2025.12
30.9
5.9
LLaMA-2 w/ SVM
Representation=Sparse...
2025.12
29.9
5.5
LLaMA-2
2025.12
29
-
LLaMA-2 w/ CG
Representation=Concept...
2025.12
23.9
4.3
Feedback
Search any
task
Search any
task