Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Interactive Social Privacy and Theory of Mind on Social Intelligence in Adversarial Dialogue (test)
Loading...
26.7
Fooling % (Hard)
GPT-5.4
5.38
10.915
16.45
21.985
Apr 13, 2026
Fooling % (Hard)
Fooling % (All)
ToM Accuracy % (Traj)
ToM Accuracy % (Step)
Average Turns
Updated 5d ago
Evaluation Results
Method
Method
Links
Fooling % (Hard)
Fooling % (All)
ToM Accuracy % (Traj)
ToM Accuracy % (Step)
Average Turns
GPT-5.4
Defender Model=GPT-5.4...
2026.04
26.7
51.1
49.8
49.9
3.53
GPT-5.4
Defender Model=GPT-5.4...
2026.04
16.7
45.3
44.7
49.9
4.2
GPT-5.4
Defender Model=GPT-5.4...
2026.04
15.6
24.7
22
45.6
4.99
GPT-5.4
Defender Model=GPT-5.4...
2026.04
6.2
38.7
40.7
46.2
4.15
Feedback
Search any
task
Search any
task
