Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Bongard Problem Solving on Bongard HOI
Loading...
91.4
Generative Accuracy
Human
50.528
61.139
71.75
82.361
Jul 10, 2025
Generative Accuracy
LSC Accuracy
Representation Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Generative Accuracy
LSC Accuracy
Representation Accuracy
Human
2025.07
91.4
-
-
LoRA (LNT)
Model=Gemma3 4B¹
2025.07
84.2
74.1
50
LoRA (Lcombined)
Model=Gemma3 4B¹
2025.07
84.2
74.1
83.2
LoRA (Lcombined)
Model=Pixtral
2025.07
79.6
63.1
77.8
LoRA (Lcombined)
Model=Phi
2025.07
79.2
71.9
82
LoRA (LNT)
Model=Phi
2025.07
78.6
71.9
63.6
LoRA (LNT)
Model=Pixtral
2025.07
78
61.6
74.9
Direct baseline
Model=Pixtral
2025.07
57.8
62.7
70.2
Direct baseline
Model=Gemma3 4B¹
2025.07
56.5
74.1
50
Direct baseline
Model=Phi
2025.07
52.1
71.9
60.5
Feedback
Search any
task
Search any
task