Share your thoughts, 1 month free Claude Pro on usSee more

Social Dialogue on SOTOPIA Interaction with GPT-4o-mini

7.53GOAL Score

GPT-4-turbo

Updated 1mo ago

Evaluation Results

Method	Links
GPT-4-turbo 2025.01		7.53	2.54
Llama-8B+BC+SDPO 2025.01		7.53	2.71
GPT-4o 2025.01		7.47	2.4
Llama-8B+BC+DMPO 2025.01		7.41	2.54
Llama-8B+BC+ETO 2025.01		7.38	2.56
Llama-8B+BC+DPO 2025.01		7.32	2.7
Llama-8B 2025.01		7.19	2.13
Llama-8B+BC 2025.01		7.18	2.59
Llama-8B+BC+Preferred-SFT 2025.01		7.18	2.52
GPT-4o-mini 2025.01		6.98	2.11
GPT-3.5-turbo 2025.01		6.67	1.84