Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-Choice on Extreme Illumination Triplet Instruct-Following Dataset
Loading...
53.13
Accuracy
Ours-7B
1.1612
14.6531
28.145
41.6369
Mar 29, 2026
Accuracy
F1 Score
Updated 2mo ago
Evaluation Results
Method
Method
Links
Accuracy
F1 Score
Ours-7B
MLLM Type=Event-Enhanc...
2026.03
53.13
85.43
Baseline-7B
MLLM Type=Event-Enhanc...
2026.03
44.71
81.95
Ours-3B
MLLM Type=Event-Enhanc...
2026.03
34.48
80.35
InternVL
MLLM Type=General Model
2026.03
31.5
75.46
Baseline-3B
MLLM Type=Event-Enhanc...
2026.03
29.86
77.04
LLaVA-7B
MLLM Type=General Model
2026.03
18.43
65.39
Q-Instruct
MLLM Type=Illumination...
2026.03
15.86
65.09
Deepseek-VL
MLLM Type=General Model
2026.03
7
55.7
EventGPT
MLLM Type=Event-specia...
2026.03
3.16
73.6
Feedback
Search any
task
Search any
task