List your company

Weather Captioning on Weather Captioning Human Evaluation Set

8.52SA ScoreWeatherTGD
5.58726.34867.117.8714Mar 23, 2026

Evaluation results

MethodLinks
WeatherTGDBackbone=Qwen3-Next-80B2026.038.528.428.688.5478
AgentVerseBackbone=Qwen3-Next-80B2026.037.326.987.187.1475
LLM-DebateBackbone=Qwen3-Next-80B2026.037.247.086.987.1276
MADBackbone=Qwen3-Next-80B2026.037.167.026.97.0277
CAMELBackbone=Qwen3-Next-80B2026.037.16.886.766.978
AutoGenBackbone=Qwen3-Next-80B2026.036.926.666.86.7879
Self-ConsistencyBackbone=Qwen3-Next-80B2026.036.886.626.946.7880
VanillaBackbone=Qwen3-Next-80B2026.035.75.485.625.682