List your company

Grid-world Navigation on FrozenLake Implicit Hidden Drift

88.8Success Rate (Source)Qwen3-30B
9.44830.04950.6571.251Jan 27, 2026

Evaluation results

MethodLinks
Qwen3-30BBackbone=Qwen3-30B, GLOVE Augmentation=false2026.0188.846.9—
Qwen3-30B + GLOVEBackbone=Qwen3-30B, GLOVE Augmentation=true2026.018547.50.6
GPT-4oBackbone=GPT-4o, GLOVE Augmentation=false2026.016051.9—
GPT-4o + GLOVEBackbone=GPT-4o, GLOVE Augmentation=true2026.0151.296.244.3
Qwen2.5-7B + GLOVEBackbone=Qwen2.5-7B, GLOVE Augmentation=true2026.0141.91002.5
Qwen2.5-7BBackbone=Qwen2.5-7B, GLOVE Augmentation=false2026.0140.697.5—
Llama3.3-70B + GLOVEBackbone=Llama3.3-70B, GLOVE Augmentation=true2026.014097.51.3
Llama3.3-70BBackbone=Llama3.3-70B, GLOVE Augmentation=false2026.0139.496.2—
Grok-3Backbone=Grok-3, GLOVE Augmentation=false2026.0135.668.1—
Grok-3 + GLOVEBackbone=Grok-3, GLOVE Augmentation=true2026.0133.898.830.7
DeepSeek-V3.2 + GLOVEBackbone=DeepSeek-V3.2, GLOVE Augmentation=true2026.0127.595.620
DeepSeek-V3.2Backbone=DeepSeek-V3.2, GLOVE Augmentation=false2026.0126.975.6—
Llama3.1-8BBackbone=Llama3.1-8B, GLOVE Augmentation=false2026.0113.152.5—
Llama3.1-8B + GLOVEBackbone=Llama3.1-8B, GLOVE Augmentation=true2026.0112.587.535