Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Captioning on W3D Safety-Critical Situation original (test)
Loading...
44
BLEU
LLada*
9.68
18.59
27.5
36.41
Nov 16, 2025
BLEU
METEOR
ROUGE
CIDEr-R
Updated 1mo ago
Evaluation Results
Method
Method
Links
BLEU
METEOR
ROUGE
CIDEr-R
LLada*
Training Regime=Full-d...
2025.11
44
38
59
1.23
FSDAM
Training Regime=Few-shot
2025.11
35
33
46
0.47
MLNet + LLaVA
Training Regime=Few-shot
2025.11
26
20
32
0.12
Qwen-VL
Training Regime=In-con...
2025.11
21
17
30
0.23
GazeXplain*
Training Regime=Full-d...
2025.11
19
29
37
0.55
Qwen-VL
Training Regime=Zero-s...
2025.11
19
21
29
0.13
LLaVA
Training Regime=Zero-s...
2025.11
13
19
11
0.1
DeepGazeI + LLaVA
Training Regime=Few-shot
2025.11
13
22
30
0.18
DeepGazeIIE + LLaVA
Training Regime=Few-shot
2025.11
11
20
32
0.13
Feedback
Search any
task
Search any
task