Alignment on Weather
83.22AlignMemPrompt
Evaluation Results
| Method | Links | |
|---|---|---|
| MemPromptLLM backbone=GPT-oss-120B2026.01 | 83.22 | |
| LLM-as-RNNLLM backbone=Gemma-3-12B2026.01 | 82.52 | |
| MemPromptLLM backbone=Gemma-3-27B2026.01 | 81.82 | |
| LLM-as-RNNLLM backbone=Gemma-3-27B2026.01 | 81.82 | |
| LLM-as-RNNLLM backbone=GPT-oss-120B2026.01 | 81.82 | |
| MemPromptLLM backbone=GPT-oss-20B2026.01 | 81.12 | |
| LLM-as-RNNLLM backbone=GPT-oss-20B2026.01 | 81.12 | |
| FHCLLM backbone=GPT-oss-120B2026.01 | 79.72 | |
| FHCLLM backbone=GPT-oss-20B2026.01 | 79.02 | |
| FHCLLM backbone=Gemma-3-27B2026.01 | 78.32 | |
| MemPromptLLM backbone=Gemma-3-12B2026.01 | 78.32 | |
| LLM-as-RNNLLM backbone=Llama-3.1-70B2026.01 | 76.22 | |
| LLM-as-RNNLLM backbone=Gemma-3-4B2026.01 | 76.22 | |
| Zero-shotLLM backbone=Gemma-3-27B2026.01 | 75.52 | |
| Zero-shotLLM backbone=GPT-oss-120B2026.01 | 74.83 | |
| FHCLLM backbone=Gemma-3-12B2026.01 | 74.83 | |
| MemPromptLLM backbone=Gemma-3-4B2026.01 | 74.83 | |
| LLM-as-RNNLLM backbone=Llama-3.1-8B2026.01 | 74.83 | |
| Zero-shotLLM backbone=GPT-oss-20B2026.01 | 74.13 | |
| MemPromptLLM backbone=Llama-3.1-70B2026.01 | 74.13 | |
| LLM-as-RNNLLM backbone=Llama-3.2-3B2026.01 | 74.13 | |
| LLM-as-RNNLLM backbone=Gemma-3-1B2026.01 | 74.13 | |
| MemPromptLLM backbone=Llama-3.1-8B2026.01 | 73.43 | |
| LLM-as-RNNLLM backbone=Llama-3.2-1B2026.01 | 73.43 | |
| Zero-shotLLM backbone=Gemma-3-12B2026.01 | 72.73 | |
| FHCLLM backbone=Llama-3.1-70B2026.01 | 72.73 | |
| FHCLLM backbone=Gemma-3-4B2026.01 | 72.73 | |
| MemPromptLLM backbone=Llama-3.2-3B2026.01 | 72.73 | |
| MemPromptLLM backbone=Gemma-3-1B2026.01 | 72.73 | |
| Zero-shotLLM backbone=Llama-3.1-70B2026.01 | 72.03 | |
| FHCLLM backbone=Llama-3.1-8B2026.01 | 72.03 | |
| MemPromptLLM backbone=Llama-3.2-1B2026.01 | 72.03 | |
| FHCLLM backbone=Llama-3.2-1B2026.01 | 71.33 | |
| FHCLLM backbone=Llama-3.2-3B2026.01 | 71.33 | |
| FHCLLM backbone=Gemma-3-1B2026.01 | 71.33 | |
| Zero-shotLLM backbone=Llama-3.1-8B2026.01 | 70.63 | |
| Zero-shotLLM backbone=Gemma-3-4B2026.01 | 70.63 | |
| Zero-shotLLM backbone=Llama-3.2-3B2026.01 | 69.23 | |
| Zero-shotLLM backbone=Gemma-3-1B2026.01 | 68.53 | |
| Zero-shotLLM backbone=Llama-3.2-1B2026.01 | 67.83 |