Agent Behavior Adaptation on FrozenLake (FL) (test)
0Loop RatioDeepSeek-V3.2
Evaluation results
| Method | Links | |
|---|---|---|
| DeepSeek-V3.2Model Type=Non-thinking Model2026.02 | 0 | |
| DeepSeek-R1Model Type=Thinking Model2026.02 | 0 | |
| Gemini 2.5 ProModel Type=Thinking Model2026.02 | 0.2 | |
| gpt-oss-120bModel Type=Thinking Model2026.02 | 0.7 | |
| Gemini 2.5 FlashModel Type=Non-thinking Model2026.02 | 1 | |
| Phi-4-reasoningModel Type=Thinking Model2026.02 | 1.4 | |
| Ministral-3-14B-InstructModel Type=Non-thinking Model2026.02 | 4.3 | |
| Glm-4-32B-0414Model Type=Non-thinking Model2026.02 | 5.1 | |
| Qwen3-30B-A3B-ThinkingModel Type=Thinking Model2026.02 | 5.3 | |
| Llama-3.1-8B-InstructModel Type=Non-thinking Model2026.02 | 5.4 | |
| Llama-3.3-70B-InstructModel Type=Non-thinking Model2026.02 | 9.5 | |
| Qwen3-4B-ThinkingModel Type=Thinking Model2026.02 | 9.8 | |
| Glm-4-9B-ChatModel Type=Non-thinking Model2026.02 | 16.7 | |
| Phi-4Model Type=Non-thinking Model2026.02 | 24.3 | |
| Qwen3-4B-InstructModel Type=Non-thinking Model2026.02 | 32 | |
| Qwen3-30B-A3B-InstructModel Type=Non-thinking Model2026.02 | 37.7 | |
| Mistral-7B-InstructModel Type=Non-thinking Model2026.02 | 63.3 |