List your company

Reasoning on LIAR Ambiguity-Augmented subset of 200 samples

69Accuracy@1DisambiguSLM
49.7654.75559.7564.745Apr 25, 2026

Evaluation results

MethodLinks
DisambiguSLM2026.046972.612.5
OPRO2026.0467.67021.4
SPO2026.0467.169.820.2
APE2026.0465.967.923.8
TextGrad2026.0465.767.823.1
CoT2026.0465.467.224.6
PromptAgent2026.0464.166.524.2
Naïve prompting2026.0463.56526.9
PromptBreeder2026.0463.265.724.9
Step-back2026.0462.865.125.3
Rephrase2026.0450.552.829.4