Grid-world Navigation on FrozenLake Explicit Structural Drift II
83.7Success Rate (Source)Llama3.3-70B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Llama3.3-70BBackbone=Llama3.3-70B, GLOVE Augmentation=false2026.01 | 83.7 | 0 | — | |
| Llama3.3-70B + GLOVEBackbone=Llama3.3-70B, GLOVE Augmentation=true2026.01 | 82.5 | 73.8 | 73.8 | |
| Qwen3-30BBackbone=Qwen3-30B, GLOVE Augmentation=false2026.01 | 80 | 13.8 | — | |
| Qwen3-30B + GLOVEBackbone=Qwen3-30B, GLOVE Augmentation=true2026.01 | 80 | 75 | 61.2 | |
| Qwen2.5-7B + GLOVEBackbone=Qwen2.5-7B, GLOVE Augmentation=true2026.01 | 78.8 | 68.8 | 68.8 | |
| GPT-4o + GLOVEBackbone=GPT-4o, GLOVE Augmentation=true2026.01 | 77.5 | 76.2 | 65 | |
| Grok-3 + GLOVEBackbone=Grok-3, GLOVE Augmentation=true2026.01 | 75 | 71.2 | 60 | |
| DeepSeek-V3.2 + GLOVEBackbone=DeepSeek-V3.2, GLOVE Augmentation=true2026.01 | 68.8 | 70 | 70 | |
| Grok-3Backbone=Grok-3, GLOVE Augmentation=false2026.01 | 60 | 11.2 | — | |
| Llama3.1-8B + GLOVEBackbone=Llama3.1-8B, GLOVE Augmentation=true2026.01 | 36.2 | 67.5 | 66.3 | |
| Qwen2.5-7BBackbone=Qwen2.5-7B, GLOVE Augmentation=false2026.01 | 6.2 | 0 | — | |
| GPT-4oBackbone=GPT-4o, GLOVE Augmentation=false2026.01 | 3.8 | 11.2 | — | |
| Llama3.1-8BBackbone=Llama3.1-8B, GLOVE Augmentation=false2026.01 | 0 | 1.2 | — | |
| DeepSeek-V3.2Backbone=DeepSeek-V3.2, GLOVE Augmentation=false2026.01 | 0 | 0 | — |