Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Preference-aligned decision making on AmbiK (test)
Loading...
84.6
Accuracy
Adaptive CLIPR
48.824
58.112
67.4
76.688
May 12, 2026
Accuracy
Updated 20d ago
Evaluation Results
Method
Method
Links
Accuracy
Adaptive CLIPR
Feedback=Adaptive
2026.05
84.6
CLIPR
Feedback=Iterative lea...
2026.05
82.6
ICL + Answers
Mode=In-context learni...
2026.05
82.4
TidyBot
2026.05
78.6
CIPHER (Sem.)
Distance Metric=Semantic
2026.05
71.7
CIPHER (Lev.)
Distance Metric=Levens...
2026.05
69.4
GATE (15-turn)
Interaction turns=15
2026.05
62.3
ICL
Mode=In-context learning
2026.05
53.8
IP
Method Identity=Intros...
2026.05
51.7
Zero-shot
Protocol=Zero-shot
2026.05
50.2
Feedback
Search any
task
Search any
task