List your company

Counterfactual Generation on IMDb

100LFRiFlip-Conf
24.49644.09863.783.302Jan 4, 2026

Evaluation results

MethodLinks
iFlip-ConfExplained Model=RoBERTa-base, Generator Model=LLaMA 3.3-70B, Feedback Type=Confidence2026.0110089.537.48
iFlip-SHAPExplained Model=RoBERTa-base, Generator Model=LLaMA 3.3-70B, Feedback Type=Feature Attribution2026.0110089.536.19
iFlip-NLExplained Model=RoBERTa-base, Generator Model=LLaMA 3.3-70B, Feedback Type=Natural Language2026.0110089.336.48
iFlipBackbone LLM=Qwen3-32B, Feedback Type=AttnLRP2026.0110085.833.19
iFlipBackbone LLM=Qwen3-32B, Feedback Type=LIME2026.0199.886.332.57
iFlipBackbone LLM=Qwen3-32B, Feedback Type=Confidence2026.0199.685.533.3
iFlipBackbone LLM=Qwen3-32B, Feedback Type=Grad×Input2026.0199.686.533.65
iFlipBackbone LLM=LLaMA3.3-70B, Feedback Type=Confidence2026.0199.687.734.74
iFlipBackbone LLM=LLaMA3.3-70B, Feedback Type=SHAP2026.0199.687.834.32
iFlipBackbone LLM=LLaMA3.3-70B, Feedback Type=Grad×Input2026.0199.687.834.8
iFlipBackbone LLM=LLaMA3.3-70B, Feedback Type=LIME2026.0199.687.734.54
iFlipBackbone LLM=LLaMA3.3-70B, Feedback Type=Natural Language2026.0199.689.536.9
iFlipBackbone LLM=Qwen3-32B, Feedback Type=SHAP2026.0199.485.733.7
iFlipBackbone LLM=LLaMA3.3-70B, Feedback Type=AttnLRP2026.0199.287.934.58
iFlipBackbone LLM=OLMo2-7B, Feedback Type=Natural Language2026.0198.482.236.47
iFlipBackbone LLM=Qwen3-32B, Feedback Type=Natural Language2026.019887.832.99
iFlipBackbone LLM=OLMo2-7B, Feedback Type=AttnLRP2026.0197.880.342.61
iFlipBackbone LLM=OLMo2-7B, Feedback Type=Grad×Input2026.0197.680.143.3
iFlipBackbone LLM=OLMo2-7B, Feedback Type=SHAP2026.0196.880.541.87
iFlipBackbone LLM=OLMo2-7B, Feedback Type=LIME2026.0196.880.552.82
iFlipBackbone LLM=OLMo2-7B, Feedback Type=Confidence2026.0196.480.941.19
FIZLEExplained Model=RoBERTa-base, Generator Model=LLaMA 3.3-70B2026.0195.686.834.88
CGGExplained Model=RoBERTa-base, Generator Model=LLaMA 3.3-70B2026.019389.254.57
FIZLEBackbone LLM=LLaMA3.3-70B2026.0192.486.834.88
Causal What-IfsBackbone LLM=LLaMA3.3-70B2026.0192.278.831.93
Causal What-IfsExplained Model=RoBERTa-base, Generator Model=LLaMA 3.3-70B2026.0190.878.832.78
Causal What-IfsBackbone LLM=OLMo2-7B2026.018980.137.97
CGGBackbone LLM=LLaMA3.3-70B2026.0188.786.952.81
CGGBackbone LLM=OLMo2-7B2026.0188.583.744.67
CGGBackbone LLM=Qwen3-32B2026.0186.387.762.47
Causal What-IfsBackbone LLM=Qwen3-32B2026.0184.882.142.03
FIZLEBackbone LLM=Qwen3-32B2026.0168.49036
BAEExplained Model=RoBERTa-base2026.0163.797.762.26
FIZLEBackbone LLM=OLMo2-7B2026.0158.558.438.71
BAE2026.0155.397.663.15
Polyjuice2026.0127.573.660.81
PolyjuiceExplained Model=RoBERTa-base2026.0127.473.660.81