List your company

Social Reasoning on Sotopia (all)

2.73Rel ScoreFT+MS
0.17160.83581.52.1642Sep 26, 2025

Evaluation results

MethodLinks
FT+MSBackbone=Qwen2.5-7B2025.092.734.47.464.86
FT+MS+Uttr (TOMA)Backbone=Qwen2.5-7B2025.092.74.777.675.05
Base+MSBackbone=Qwen2.5-7B2025.092.474.457.34.74
FT+UttrBackbone=Qwen2.5-7B2025.092.424.787.434.88
FT+MSBackbone=Qwen2.5-3B2025.092.373.816.694.29
FT+MS+Uttr (TOMA)Backbone=Llama3.1-8B2025.092.375.617.485.15
FT+MS+Uttr (TOMA)Backbone=Qwen2.5-3B2025.092.184.226.844.41
BaseBackbone=Qwen2.5-7B2025.092.074.547.264.62
FT+UttrBackbone=Qwen2.5-3B2025.091.924.016.64.18
Base+MSBackbone=Qwen2.5-3B2025.091.543.485.933.65
FT+MSBackbone=Llama3.1-8B2025.091.494.76.464.22
FT+UttrBackbone=Llama3.1-8B2025.091.285.186.884.45
Base+MSBackbone=Llama3.1-8B2025.091.25.376.674.41
BaseBackbone=Qwen2.5-3B2025.090.973.295.253.17
BaseBackbone=Llama3.1-8B2025.090.275.096.113.82