List your company

Counterfactual Text Generation on CLASSE

57.5Sim ScoreGumbel Machine
26.8234.78542.7550.715May 26, 2026

Evaluation results

MethodLinks
Gumbel MachineModel=Qwen3-4B-Instruct2026.0557.556.6
Identify-and-ReplaceModel=Claude Sonnet 4.62026.0555.443.4
Identify-and-ReplaceModel=LLaMA-3.1-8B-Instruct2026.0549.810.3
Identify-and-ReplaceModel=Claude Haiku 4.52026.0545.138.8
Gumbel MachineModel=LLaMA-3.1-8B-Instruct2026.0539.166.8
Gumbel MachineModel=LLaMA-3.1-8B-Instruct2026.0539.166.8
Minimal-EditModel=Claude Sonnet 4.62026.0537.634.4
In-Context LearningModel=Claude Sonnet 4.62026.0535.960.4
Minimal-EditModel=LLaMA-3.1-8B-Instruct2026.0531.418.3
In-Context LearningModel=LLaMA-3.1-8B-Instruct2026.0530.832.7
In-Context LearningModel=Claude Haiku 4.52026.0528.131.2
Minimal-EditModel=Claude Haiku 4.52026.052814.4