List your company

Binary Classification on LIAR Open

84.21Accuracyw/ Intervention
43.514854.079964.64575.2101Feb 12, 2024

Evaluation results

MethodLinks
w/ InterventionLLM=Llama2 (13B)2024.0284.2184.03
TELLERLLM=Llama2 (13B)2024.0282.5282.37
w/ InterventionLLM=Llama2 (7B)2024.0281.9381.84
Few-shotLLM=GPT-3.5-turbo2024.0281.0281
w/ InterventionLLM=FLAN-T5-xxl (11B)2024.0280.7380.41
TELLERLLM=FLAN-T5-xxl (11B)2024.0280.2479.85
TELLERLLM=Llama2 (7B)2024.0279.9479.8
w/ InterventionLLM=FLAN-T5-xl (3B)2024.0279.3479.07
TELLERLLM=FLAN-T5-xl (3B)2024.0278.7578.55
TELLERLLM=FLAN-T5-large (780M)2024.0277.7677.32
w/ InterventionLLM=FLAN-T5-large (780M)2024.0277.4677.14
DirectLLM=GPT-3.5-turbo2024.0276.2774.21
DirectLLM=FLAN-T5-xl (3B)2024.0275.9775.67
DirectLLM=FLAN-T5-xxl (11B)2024.0275.1775.15
Few-shot COTLLM=GPT-3.5-turbo2024.0274.4876.21
DirectLLM=FLAN-T5-large (780M)2024.0274.3873.84
Zero-shot COTLLM=GPT-3.5-turbo2024.0272.4970.83
DirectLLM=Llama2 (7B)2024.0272.2969.63
DirectLLM=Llama2 (13B)2024.0269.3163.77
Few-shot LogicLLM=GPT-3.5-turbo2024.0261.6760.92
DirectLLM=FLAN-T5-base (250M)2024.0261.4761.43
DirectLLM=FLAN-T5-small (80M)2024.0245.0832.41