Answer Sentence Selection on PrivacyQA 1.0 (test) benchmark leaderboard · SOTA2 Research

Answer Sentence Selection on PrivacyQA 1.0 (test)

0.611SAEGPT-4o-mini Multi-agent

Evaluation Results

Method	Links
GPT-4o-mini Multi-agent2025.06		0.611	0.595	0.596	0.602	0.592	0.594	0.598	0.005	0.019
GPT-4o-mini2025.06		0.605	0.573	0.562	0.555	0.547	0.547	0.565	0.016	0.058
GPT-4o-mini Multi-agent2025.06		0.601	0.588	0.578	0.587	0.592	0.576	0.587	0.007	0.025
DeepSeek-R1 Multi-agent2025.06		0.582	0.579	0.583	0.579	0.566	0.573	0.577	0.005	0.017
DeepSeek-R12025.06		0.581	0.549	0.547	0.517	0.556	0.541	0.549	0.014	0.064
Llama 3.1 Multi-agent2025.06		0.555	0.525	0.523	0.529	0.522	0.528	0.53	0.008	0.033
Llama 3.1 Multi-agent2025.06		0.549	0.527	0.52	0.524	0.523	0.526	0.528	0.007	0.029
Llama 3.12025.06		0.546	0.463	0.469	0.448	0.485	0.446	0.476	0.026	0.1
DeepSeek-R1 Multi-agent2025.06		0.533	0.606	0.585	0.581	0.557	0.569	0.572	0.019	0.073
DeepSeek-R12025.06		0.532	0.51	0.547	0.529	0.532	0.512	0.527	0.011	0.037
Llama 3.12025.06		0.469	0.349	0.37	0.325	0.356	0.336	0.368	0.035	0.144
GPT-4o-mini2025.06		0.394	0.344	0.332	0.329	0.312	0.301	0.335	0.022	0.093