Weather Report Generation on WeatherTGD Dataset
8.68LLM SA ScoreWeatherTGD
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| WeatherTGDLLM Backbone=Qwen3-Next, Tok.=3.52026.03 | 8.68 | 8.56 | 8.82 | 8.69 | 8.52 | 8.42 | 8.68 | 8.54 | 48.2 | 54.2 | 85.8 | |
| WeatherTGDLLM Backbone=MiniMax-01, Tok.=3.52026.03 | 8.52 | 8.38 | 8.68 | 8.53 | 8.36 | 8.22 | 8.52 | 8.37 | 46.8 | 52.8 | 84.5 | |
| WeatherTGDLLM Backbone=Average, Tok.=3.52026.03 | 8.48 | 8.37 | 8.64 | 8.5 | 8.32 | 8.22 | 8.49 | 8.34 | 46.7 | 52.7 | 84.5 | |
| WeatherTGDLLM Backbone=DeepSeek-V3, Tok.=3.52026.03 | 8.24 | 8.18 | 8.42 | 8.28 | 8.08 | 8.02 | 8.26 | 8.12 | 45.2 | 51.2 | 83.2 | |
| AVerseLLM Backbone=Qwen3-Next, Tok.=3.82026.03 | 7.45 | 7.12 | 7.32 | 7.3 | 7.32 | 6.98 | 7.18 | 7.14 | 41.8 | 48.5 | 81.5 | |
| DebateLLM Backbone=Qwen3-Next, Tok.=4.52026.03 | 7.38 | 7.22 | 7.12 | 7.24 | 7.24 | 7.08 | 6.98 | 7.12 | 41.2 | 47.8 | 80.8 | |
| MADLLM Backbone=Qwen3-Next, Tok.=4.22026.03 | 7.28 | 7.14 | 7.04 | 7.15 | 7.16 | 7.02 | 6.9 | 7.02 | 40.5 | 47.2 | 80.5 | |
| CAMELLLM Backbone=Qwen3-Next, Tok.=3.22026.03 | 7.24 | 7.02 | 6.92 | 7.06 | 7.1 | 6.88 | 6.76 | 6.9 | 39.8 | 46.2 | 79.8 | |
| AVerseLLM Backbone=MiniMax-01, Tok.=3.82026.03 | 7.18 | 6.82 | 7.02 | 7.01 | 7.04 | 6.68 | 6.88 | 6.86 | 39.8 | 46.2 | 80.2 | |
| AVerseLLM Backbone=Average, Tok.=3.82026.03 | 7.18 | 6.83 | 7.02 | 7.01 | 7.04 | 6.68 | 6.88 | 6.86 | 39.9 | 46.5 | 80.2 | |
| DebateLLM Backbone=MiniMax-01, Tok.=4.52026.03 | 7.12 | 6.94 | 6.82 | 6.96 | 6.98 | 6.8 | 6.68 | 6.8 | 39.5 | 45.8 | 79.5 | |
| DebateLLM Backbone=Average, Tok.=4.52026.03 | 7.12 | 6.95 | 6.83 | 6.97 | 6.97 | 6.8 | 6.69 | 6.82 | 39.5 | 45.9 | 79.5 | |
| AutoGenLLM Backbone=Qwen3-Next, Tok.=2.82026.03 | 7.08 | 6.82 | 6.96 | 6.95 | 6.92 | 6.66 | 6.8 | 6.78 | 38.8 | 44.8 | 78.8 | |
| MADLLM Backbone=MiniMax-01, Tok.=4.22026.03 | 7.04 | 6.88 | 6.74 | 6.89 | 6.9 | 6.74 | 6.58 | 6.74 | 38.8 | 45.2 | 79.2 | |
| MADLLM Backbone=Average, Tok.=4.22026.03 | 7.03 | 6.88 | 6.75 | 6.89 | 6.89 | 6.75 | 6.61 | 6.75 | 38.8 | 45.3 | 79.2 | |
| SCLLM Backbone=Qwen3-Next, Tok.=5.02026.03 | 7.02 | 6.76 | 7.08 | 6.95 | 6.88 | 6.62 | 6.94 | 6.78 | 38.2 | 44.2 | 78.2 | |
| CAMELLLM Backbone=MiniMax-01, Tok.=3.22026.03 | 6.98 | 6.72 | 6.64 | 6.78 | 6.82 | 6.56 | 6.48 | 6.62 | 38.2 | 44.5 | 78.6 | |
| CAMELLLM Backbone=Average, Tok.=3.22026.03 | 6.98 | 6.73 | 6.65 | 6.79 | 6.82 | 6.57 | 6.49 | 6.62 | 38.2 | 44.5 | 78.6 | |
| AVerseLLM Backbone=DeepSeek-V3, Tok.=3.82026.03 | 6.92 | 6.54 | 6.72 | 6.73 | 6.76 | 6.38 | 6.58 | 6.57 | 38.2 | 44.8 | 78.8 | |
| DebateLLM Backbone=DeepSeek-V3, Tok.=4.52026.03 | 6.85 | 6.68 | 6.56 | 6.7 | 6.68 | 6.52 | 6.42 | 6.55 | 37.8 | 44.2 | 78.2 | |
| AutoGenLLM Backbone=MiniMax-01, Tok.=2.82026.03 | 6.82 | 6.54 | 6.68 | 6.68 | 6.66 | 6.38 | 6.52 | 6.5 | 37.2 | 43.2 | 77.8 | |
| AutoGenLLM Backbone=Average, Tok.=2.82026.03 | 6.81 | 6.55 | 6.69 | 6.68 | 6.65 | 6.39 | 6.53 | 6.51 | 37.3 | 43.4 | 77.8 | |
| MADLLM Backbone=DeepSeek-V3, Tok.=4.22026.03 | 6.78 | 6.62 | 6.48 | 6.63 | 6.62 | 6.48 | 6.34 | 6.48 | 37.2 | 43.5 | 77.8 | |
| SCLLM Backbone=Average, Tok.=5.02026.03 | 6.75 | 6.49 | 6.8 | 6.68 | 6.59 | 6.35 | 6.65 | 6.52 | 36.7 | 42.8 | 77.2 | |
| SCLLM Backbone=MiniMax-01, Tok.=5.02026.03 | 6.74 | 6.48 | 6.78 | 6.67 | 6.58 | 6.34 | 6.62 | 6.52 | 36.8 | 42.8 | 77.2 | |
| CAMELLLM Backbone=DeepSeek-V3, Tok.=3.22026.03 | 6.72 | 6.45 | 6.38 | 6.52 | 6.54 | 6.28 | 6.22 | 6.35 | 36.5 | 42.8 | 77.4 | |
| AutoGenLLM Backbone=DeepSeek-V3, Tok.=2.82026.03 | 6.54 | 6.28 | 6.42 | 6.41 | 6.38 | 6.12 | 6.26 | 6.24 | 35.8 | 42.1 | 76.8 | |
| SCLLM Backbone=DeepSeek-V3, Tok.=5.02026.03 | 6.48 | 6.22 | 6.54 | 6.41 | 6.32 | 6.08 | 6.38 | 6.26 | 35.2 | 41.5 | 76.2 | |
| VanillaLLM Backbone=Qwen3-Next, Tok.=1.02026.03 | 5.85 | 5.62 | 5.74 | 5.74 | 5.7 | 5.48 | 5.62 | 5.6 | 34.2 | 40.8 | 75.8 | |
| VanillaLLM Backbone=MiniMax-01, Tok.=1.02026.03 | 5.68 | 5.42 | 5.58 | 5.56 | 5.52 | 5.28 | 5.44 | 5.42 | 32.8 | 39.8 | 75.2 | |
| VanillaLLM Backbone=Average, Tok.=1.02026.03 | 5.65 | 5.41 | 5.56 | 5.54 | 5.5 | 5.27 | 5.43 | 5.4 | 32.7 | 39.7 | 75.1 | |
| VanillaLLM Backbone=DeepSeek-V3, Tok.=1.02026.03 | 5.42 | 5.18 | 5.36 | 5.32 | 5.28 | 5.04 | 5.22 | 5.18 | 31.2 | 38.5 | 74.2 |