List your company

Question Answering on ECQA, QASC, OBQA, ARC

79.49ECQA Accuracy+ RLMEC
34.551646.218357.88569.5517Jan 11, 2024

Evaluation results

MethodLinks
+ RLMECModel Parameters=13B, Base Model=SFT LLM2024.0179.4964.1565.665.1968.61
+ DPOModel Parameters=13B, Base Model=SFT LLM2024.0178.2661.4562.263.866.43
+ RFT w/ TDModel Parameters=13B, Base Model=SFT LLM2024.0176.7161.5661.864.1466.05
+ RFT w/ GTModel Parameters=13B, Base Model=SFT LLM2024.0176.6660.3763.463.1765.9
+ RFT w/ RDModel Parameters=13B, Base Model=SFT LLM2024.0176.6262.263.263.1766.3
+ CoHModel Parameters=13B, Base Model=SFT LLM2024.0176.6260.3759.863.9365.18
+ Vanilla PPOModel Parameters=13B, Base Model=SFT LLM2024.0176.3457.9961.862.2964.61
SFT LLMModel Parameters=13B2024.0176.1259.460.862.4664.7
+ SFTModel Parameters=13B, Base Model=SFT LLM2024.0175.8957.8763.462.564.92
+ RFTModel Parameters=13B, Base Model=SFT LLM2024.0175.7160.486164.0665.31
+ RLMECModel Parameters=7B, Base Model=SFT LLM2024.0173.6659.556.858.562.12
+ RFT w/ TDModel Parameters=7B, Base Model=SFT LLM2024.0173.1158.2154.257.5360.76
+ Vanilla PPOModel Parameters=7B, Base Model=SFT LLM2024.0172.8850.2243.456.2755.69
+ RFT w/ GTModel Parameters=7B, Base Model=SFT LLM2024.0172.4758.5353.657.1160.43
+ RFT w/ RDModel Parameters=7B, Base Model=SFT LLM2024.0172.4759.2954.657.0360.85
+ DPOModel Parameters=7B, Base Model=SFT LLM2024.0172.4758.5355.455.2660.42
+ RFTModel Parameters=7B, Base Model=SFT LLM2024.0172.2458.6455.257.1560.81
SFT LLMModel Parameters=7B2024.0171.8855.45256.2758.89
+ CoHModel Parameters=7B, Base Model=SFT LLM2024.0171.0654.8651.456.6158.48
+ PPO A2CModel Parameters=7B, Base Model=SFT LLM2024.0170.8355.0852.456.0258.58
+ SFTModel Parameters=7B, Base Model=SFT LLM2024.0170.6555.9451.656.9958.8
+ FIGAModel Parameters=7B, Base Model=SFT LLM2024.0169.8352.485146.2154.88
LLaMA 2Model Parameters=13B2024.0161.5345.4657.964.3157.3
+ FIGAModel Parameters=13B, Base Model=SFT LLM2024.0161.2160.2652.846.3455.15
LLaMA 2Model Parameters=7B2024.0155.9739.7448.452.4849.15
WizardLMModel Parameters=13B2024.0152.640.9352.358.9651.2
VicunaModel Parameters=13B2024.0150.1439.9648.453.748.05
VicunaModel Parameters=7B2024.0149.8232.1846.451.5244.98
WizardLMModel Parameters=7B2024.0136.2818.6827.846.5932.34