List your company

Truthfulness on TruthfulQA (Reward, TR, IR, Div)

-1.8RewardSEA
-6.3136-5.1418-3.97-2.7982May 26, 2025

Evaluation results

MethodLinks
SEAModel Backbone=LLaMA-3.2-Instruct (1B)2025.05-1.8899987
CBSModel Backbone=LLaMA-3.2-Base (3B)2025.05-2.596710087
SEAModel Backbone=LLaMA-3-Base (8B)2025.05-2.66769987
BoN-64Model Backbone=LLaMA-3.2-Base (3B)2025.05-2.737410085
SEAModel Backbone=LLaMA-3.2-Base (3B)2025.05-2.938010091
RSModel Backbone=LLaMA-3-Base (8B)2025.05-3679984
BoN-64Model Backbone=LLaMA-3.2-Instruct (1B)2025.05-3.01779883
BoN-64Model Backbone=LLaMA-3-Base (8B)2025.05-3.1729885
CBSModel Backbone=LLaMA-3.2-Instruct (1B)2025.05-3.17759785
CBSModel Backbone=LLaMA-3-Base (8B)2025.05-3.18649883
BoN-32Model Backbone=LLaMA-3.2-Base (3B)2025.05-3.25739887
RSModel Backbone=LLaMA-3.2-Base (3B)2025.05-3.376610086
BoN-32Model Backbone=LLaMA-3.2-Instruct (1B)2025.05-3.44799784
RSModel Backbone=LLaMA-3.2-Instruct (1B)2025.05-3.54869981
BoN-32Model Backbone=LLaMA-3-Base (8B)2025.05-3.55689785
SEAModel Backbone=LLaMA-3.2-Base (1B)2025.05-3.647810090
CBSModel Backbone=LLaMA-3.2-Base (1B)2025.05-3.956710086
BoN-64Model Backbone=LLaMA-3.2-Base (1B)2025.05-4.07789981
BoN-8Model Backbone=LLaMA-3.2-Base (3B)2025.05-4.08769987
SFTModel Backbone=LLaMA-3.2-Base (3B)2025.05-4.17649889
SFTModel Backbone=LLaMA-3-Base (8B)2025.05-4.236210086
BoN-32Model Backbone=LLaMA-3.2-Base (1B)2025.05-4.397710082
BoN-8Model Backbone=LLaMA-3.2-Instruct (1B)2025.05-4.39769583
BoN-8Model Backbone=LLaMA-3-Base (8B)2025.05-4.48709785
ARGSModel Backbone=LLaMA-3.2-Base (3B)2025.05-4.59649878
ARGSModel Backbone=LLaMA-3-Base (8B)2025.05-4.68738782
SFTModel Backbone=LLaMA-3.2-Instruct (1B)2025.05-4.87729585
ARGSModel Backbone=LLaMA-3.2-Instruct (1B)2025.05-4.87729953
RSModel Backbone=LLaMA-3.2-Base (1B)2025.05-5.26669883
ARGSModel Backbone=LLaMA-3.2-Base (1B)2025.05-5.31559776
BoN-8Model Backbone=LLaMA-3.2-Base (1B)2025.05-5.64759884
SFTModel Backbone=LLaMA-3.2-Base (1B)2025.05-6.14599886