List your company

Counterfactual Text Generation on DREsS

59.3Similarity ScoreGumbel Machine
29.34837.12444.952.676May 26, 2026

Evaluation results

MethodLinks
Gumbel MachineModel=Qwen3-4B-Instruct2026.0559.331.5
Identify-and-ReplaceModel=Claude Sonnet 4.62026.0547.740.5
Gumbel MachineModel=LLaMA-3.1-8B-Instruct2026.0547.740.4
Gumbel MachineModel=LLaMA-3.1-8B-Instruct2026.0547.740.4
In-Context LearningModel=Claude Sonnet 4.62026.0541.848.4
Identify-and-ReplaceModel=Claude Haiku 4.52026.0538.444.1
Minimal-EditModel=Claude Sonnet 4.62026.0537.543.9
In-Context LearningModel=LLaMA-3.1-8B-Instruct2026.053326.4
In-Context LearningModel=Claude Haiku 4.52026.0531.412.5
Minimal-EditModel=Claude Haiku 4.52026.0531.29.3
Identify-and-ReplaceModel=LLaMA-3.1-8B-Instruct2026.0530.832.5
Minimal-EditModel=LLaMA-3.1-8B-Instruct2026.0530.518.8