List your company

Pointwise binary classification of political sentiment on D(NA) high-confidence

71.2Macro F1 ScoreEnsemble 1
69.01669.58370.1570.717Feb 20, 2026

Evaluation results

MethodLinks
Ensemble 12026.0271.271.472
Ensemble 22026.0271.271.472
Ensemble 32026.0271.271.472
GPT 4o2026.0271.271.472
DeepSeek-V32026.0271.271.472
Grok 22026.0271.271.472
Gemini 2.5 Flash2026.0271.271.472
GPT 4.52026.0271.271.472
Llama 4 Maverick2026.0271.271.472
Gemini 1.5 Pro2026.0271.271.472
Qwen 32026.0271.271.472
Claude 3.7 Sonnet2026.0271.271.472
GPT 3.5 Turbo2026.0269.169.170.5