Goal Prediction on FrozenLake ID
100Classification AccuracyT-CoT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| T-CoTSize=4, Backbone=GPT-4o2026.06 | 100 | 83.3 | |
| SVoToSize=4, Backbone=Anole2026.06 | 100 | 86.7 | |
| SVoTpSize=4, Backbone=Anole2026.06 | 100 | 83.3 | |
| SVoToSize=6, Backbone=Anole2026.06 | 93.3 | 80 | |
| SVoTpSize=6, Backbone=Anole2026.06 | 93.3 | 83.3 | |
| SVoTpSize=5, Backbone=Anole2026.06 | 91.7 | 86.7 | |
| T-CoTSize=5, Backbone=GPT-4o2026.06 | 90 | 80 | |
| SVoToSize=5, Backbone=Anole2026.06 | 90 | 83.3 | |
| MVoTSize=4, Backbone=Anole2026.06 | 86.7 | 46.7 | |
| T-CoTSize=6, Backbone=GPT-4o2026.06 | 86.7 | 60 | |
| T-CoTSize=4, Backbone=Anole2026.06 | 75 | 33.3 | |
| T-CoTSize=5, Backbone=Anole2026.06 | 73.3 | 33.3 | |
| MVoTSize=5, Backbone=Anole2026.06 | 73.3 | 43.3 | |
| MVoTSize=6, Backbone=Anole2026.06 | 63.3 | 40 | |
| T-CoTSize=6, Backbone=Anole2026.06 | 50 | 36.7 |