List your company

Best feasible prompt identification on CNN/DailyMail (test)

0.172Average Soft Constrained RewardMLP-CSR
-0.006880.039560.0860.13244May 14, 2026

Evaluation results

MethodLinks
MLP-CSRBackbone=Gemma-7B, K=50, b=82026.050.172
CSRBackbone=Gemma-7B, K=50, b=102026.050.167
CSRBackbone=Gemma-7B, K=30, b=102026.050.166
CSRBackbone=Llama3-8B, K=50, b=102026.050.165
MLP-CSRBackbone=Gemma-7B, K=30, b=52026.050.164
MLP-CSRBackbone=Gemma-7B, K=30, b=82026.050.164
MLP-CSRBackbone=Gemma-7B, K=30, b=102026.050.163
MLP-CSRBackbone=Gemma-7B, K=50, b=52026.050.162
CSRBackbone=Gemma-7B, K=50, b=52026.050.161
CSRBackbone=Gemma-7B, K=50, b=82026.050.161
MLP-CSRBackbone=Llama3-8B, K=50, b=82026.050.16
MLP-CSRBackbone=Llama3-8B, K=50, b=102026.050.16
CSRBackbone=Llama3-8B, K=100, b=102026.050.16
MLP-CSRBackbone=Llama3-8B, K=100, b=102026.050.16
CSRBackbone=Llama3-8B, K=30, b=32026.050.157
CSRBackbone=Llama3-8B, K=50, b=52026.050.157
CSRBackbone=Llama3-8B, K=50, b=82026.050.155
MLP-CSRBackbone=Llama3-8B, K=50, b=52026.050.155
CSRBackbone=Llama3-8B, K=30, b=102026.050.154
MLP-CSRBackbone=Llama3-8B, K=30, b=102026.050.154
CSRBackbone=Llama3-8B, K=50, b=32026.050.154
MLP-CSRBackbone=Llama3-8B, K=100, b=52026.050.154
MLP-CSRBackbone=Llama3-8B, K=100, b=82026.050.154
CSRBackbone=Gemma-7B, K=30, b=82026.050.153
MLP-CSRBackbone=Gemma-7B, K=50, b=102026.050.153
CSRBackbone=Llama3-8B, K=100, b=82026.050.151
CSRBackbone=Llama3-8B, K=30, b=82026.050.148
MLP-CSRBackbone=Gemma-7B, K=50, b=32026.050.146
MLP-CSRBackbone=Llama3-8B, K=50, b=32026.050.144
MLP-CSRBackbone=Llama3-8B, K=30, b=32026.050.142
MLP-CSRBackbone=Llama3-8B, K=30, b=82026.050.142
MLP-CSRBackbone=Gemma-7B, K=30, b=32026.050.138
CSRBackbone=Llama3-8B, K=100, b=52026.050.138
CSRBackbone=Llama3-8B, K=30, b=52026.050.136
CSRBackbone=Gemma-7B, K=30, b=52026.050.134
MLP-CSRBackbone=Llama3-8B, K=100, b=32026.050.134
MLP-CSRBackbone=Llama3-8B, K=30, b=52026.050.126
CSRBackbone=Llama3-8B, K=100, b=32026.050.123
CSRBackbone=Gemma-7B, K=50, b=32026.050.091
CSRBackbone=Gemma-7B, K=100, b=32026.050.088
CSRBackbone=Gemma-7B, K=30, b=32026.050.081
MLP-CSRBackbone=Gemma-7B, K=100, b=102026.050.072
CSRBackbone=Gemma-7B, K=100, b=82026.050.062
UniformBackbone=Gemma-7B, K=100, b=52026.050.045
MLP-CSRBackbone=Gemma-7B, K=100, b=52026.050.045
UniformBackbone=Llama3-8B, K=100, b=52026.050.044
UniformBackbone=Gemma-7B, K=30, b=52026.050.032
CSRBackbone=Gemma-7B, K=100, b=102026.050.032
MLP-CSRBackbone=Gemma-7B, K=100, b=32026.050.032
MLP-CSRBackbone=Gemma-7B, K=100, b=82026.050.032
UniformBackbone=Gemma-7B, K=30, b=32026.050.021
UniformBackbone=Gemma-7B, K=30, b=82026.050.021
UniformBackbone=Gemma-7B, K=30, b=102026.050.021
UniformBackbone=Gemma-7B, K=50, b=52026.050.02
UniformBackbone=Gemma-7B, K=50, b=82026.050.02
UniformBackbone=Llama3-8B, K=100, b=32026.050.019
UniformBackbone=Llama3-8B, K=100, b=82026.050.018
UniformBackbone=Llama3-8B, K=100, b=102026.050.018
UniformBackbone=Llama3-8B, K=50, b=52026.050.016
UniformBackbone=Gemma-7B, K=100, b=32026.050.015
UniformBackbone=Gemma-7B, K=100, b=102026.050.015
CSRBackbone=Gemma-7B, K=100, b=52026.050.015
UniformBackbone=Llama3-8B, K=50, b=32026.050.015
UniformBackbone=Llama3-8B, K=50, b=102026.050.01
UniformBackbone=Llama3-8B, K=30, b=82026.050.008
UniformBackbone=Llama3-8B, K=50, b=82026.050.008
UniformBackbone=Gemma-7B, K=50, b=32026.050.007
UniformBackbone=Gemma-7B, K=50, b=102026.050
UniformBackbone=Gemma-7B, K=100, b=82026.050
UniformBackbone=Llama3-8B, K=30, b=32026.050
UniformBackbone=Llama3-8B, K=30, b=52026.050
UniformBackbone=Llama3-8B, K=30, b=102026.050