Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Domain-Incremental Learning on DomainNet 365 classes (test)
Loading...
67.8
Accuracy
S-liPrompts
39.096
46.548
54
61.452
Jul 5, 2023
Accuracy
Relative Error Rate (inc. Phase 1)
Relative Error Rate (no Phase 1)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Relative Error Rate (inc. Phase 1)
Relative Error Rate (no Phase 1)
S-liPrompts
Backbone=CLIP
2023.07
67.8
-
-
RanPAC (Ours (Algorithm 1))
Backbone=ViT B/16
2023.07
66.6
25
23
RanPAC (No Phase 1)
2023.07
64.7
-
-
RanPAC (No RPs)
2023.07
55.2
-
-
RanPAC (No RPs or Phase 1)
2023.07
54.1
-
-
RanPAC (NCM with Phase 1)
2023.07
51.3
-
-
S-iPrompts
Backbone=ViT B/16
2023.07
50.6
-
-
ADaM
Backbone=ViT B/16
2023.07
50.3
-
-
RanPAC (NCM only)
2023.07
46.4
-
-
L2P
Backbone=ViT B/16
2023.07
40.2
-
-
Feedback
Search any
task
Search any
task