Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Embodied Question Answering on OpenEQA v1 (test)
Loading...
85.1
Score
Human Agent
32.372
46.061
59.75
73.439
Dec 5, 2024
Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
Human Agent
agent_type=human
2024.12
85.1
Socratic LLMs w/ Frame Captions
zero-shot=true
2024.12
38.1
TANGO
zero-shot=true
2024.12
37.2
Blind LLMs
zero-shot=true
2024.12
35.5
Socratic LLMs w/ Scene-Graph Captions
zero-shot=true
2024.12
34.4
Feedback
Search any
task
Search any
task
