Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Behavior decoding on Flint (held-out sessions)

84.3R2 Score

REALM-bi

-0.14821.77643.765.624May 14, 2026
Updated 19d ago

Evaluation Results

MethodLinks
2026.05
84.3
2026.05
79.7
2026.05
79.7
2026.05
78.9
2026.05
78.6
2026.05
78.5
2026.05
78.5
2026.05
77.7
2026.05
77.1
2026.05
76.9
2026.05
75.4
2026.05
75.2
2026.05
74.1
2026.05
73
2026.05
71.3
2026.05
70.9
2026.05
70.9
2026.05
70.6
2026.05
70.3
2026.05
70.3
2026.05
70
2026.05
69.2
2026.05
68.9
2026.05
68.9
2026.05
68.9
2026.05
68.5
2026.05
68.1
2026.05
66.7
2026.05
66.6
2026.05
65.1
2026.05
64.9
2026.05
64.4
2026.05
64.1
2026.05
64.1
2026.05
62.4
2026.05
61.6
2026.05
61.3
2026.05
60.2
2026.05
59.1
2026.05
58.7
2026.05
57.7
2026.05
57.6
2026.05
56.7
2026.05
56.4
2026.05
55.1
2026.05
54.6
2026.05
54.3
2026.05
40
2026.05
35.5
2026.05
33
2026.05
30.4
2026.05
29.6
2026.05
23
2026.05
22
2026.05
21.3
2026.05
20.1
2026.05
20
2026.05
18.5
2026.05
16.7
2026.05
13.2
2026.05
11
2026.05
3.6
2026.05
3.1