List your company

Commonsense Question Answering on CommonSenseQA

88.9AccuracyHUMAN
7.05228.30149.5570.799Mar 24, 2021Jan 16, 2022Nov 11, 2022Sep 6, 2023Jun 30, 2024Apr 25, 2025Feb 18, 2026

Evaluation results

MethodLinks
HUMAN2021.0388.9
EPR-KMeansModel=Qwen2.5-72B, Shots=1282025.0687.96
BGE-KMeansModel=Qwen2.5-72B, Shots=1282025.0687.71
CLGModel=Qwen2.5-72B, Shots=1282025.0687.71
Best-of-NModel=Qwen2.5-72B, Shots=1282025.0687.55
RandomModel=Qwen2.5-72B, Shots=1282025.0687.47
BM25-MajorModel=Qwen2.5-72B, Shots=1282025.0686.65
Llama 3 405BParameters=405B2024.0785.8
Latent-BayesianModel=Qwen2.5-72B, Shots=1282025.0685.75
QWEN-3-30B MOEPARAMS=30B, Variant=FCSD2025.1184.9
QWEN-3-30B MOEPARAMS=30B, Variant=ORIGINAL2025.1184.8
RandomModel=Llama3-70B, Shots=1282025.0684.73
Latent-BayesianModel=Llama3-70B, Shots=1282025.0684.6
CLGModel=Llama3-70B, Shots=1282025.0684.52
LLAMA-4-SCOUTPARAMS=109B, Variant=ORIGINAL2025.1184.4
Best-of-NModel=Llama3-70B, Shots=1282025.0684.19
Llama 3 70BParameters=70B2024.0784.1
BGE-KMeansModel=Llama3-70B, Shots=1282025.0683.78
EPR-KMeansModel=Llama3-70B, Shots=1282025.0683.62
Mixtral 8×22BParameters=8×22B2024.0782.4
LLAMA-4-SCOUTPARAMS=109B, Variant=FCSD2025.1182.4
TAIAModel=Qwen1.5-32B, Infer Mode=TAIA2024.0582.31
BM25-MajorModel=Llama3-70B, Shots=1282025.0682.23
BaseModel=Qwen1.5-32B, Infer Mode=Base2024.0580.92
UNICORN2021.0379.3
UNIFIEDQA-11B2021.0379.1
SDPO2026.0278.71
Qwen3-8B2026.0278.46
BaseModel=Qwen1.5-7B, Infer Mode=Base2024.0578.3
T5-11B2021.0378.1
LoRAModel=Qwen1.5-32B, Infer Mode=LoRA2024.0577.81
BaseModel=Qwen1.5-14B, Infer Mode=Base2024.0577.72
LoRAModel=Qwen1.5-7B, Infer Mode=LoRA2024.0577.31
TAIAModel=Qwen1.5-7B, Infer Mode=TAIA2024.0577.31
TAIAModel=Qwen1.5-14B, Infer Mode=TAIA2024.0577.31
Llama 3 8BParameters=8B2024.0775
LoRAModel=Qwen1.5-14B, Infer Mode=LoRA2024.0574.45
Gemma 7BParameters=7B2024.0774.4
ROBERTA-LARGE2021.0372.1
LLAMA-4-SCOUTPARAMS=109B, Variant=SFT2025.1172
Mistral 7BParameters=7B2024.0771.2
DEEPSEEK-V2-LITEPARAMS=16B, Variant=FCSD2025.1170.1
DEEPSEEK-V2-LITEPARAMS=16B, Variant=ORIGINAL2025.1169.1
QWEN-3-30B MOEPARAMS=30B, Variant=SFT2025.1168.9
DEEPSEEK-V2-LITEPARAMS=16B, Variant=SFT2025.1165.5
OLTQAConfiguration=Full2023.0557.08
OLTQAAblation=without Pm2023.0556.91
OLTQAKnowledge Distillation=Back KD2023.0556.73
OLTQAKnowledge Distillation=Static MKD2023.0556.1
OLTQAKnowledge Distillation=without MKD2023.0555.61
OLTQARetriever=EPR2023.0554.95
Muppet2023.0553.79
Hyperformer++2023.0553.24
OLTQARe-ranker=excluded2023.0552.83
OLTQARetriever=BM252023.0552.66
OLTQAAblation=without Pk2023.0552.5
UnifiedQA2023.0551.92
ProQA2023.0551.52
EPR2023.0545.3
LTVPrompt=Zero-shot2025.0241
+ LTVprompt=zero-shot2025.0241
Pre-editedBackbone=LLaMA2, Number of Edits=8000, Edit Dataset=zsRE2026.0132.84
HORSEBackbone=LLaMA2, Number of Edits=8000, Edit Dataset=zsRE2026.0129.87
MALMENBackbone=LLaMA2, Number of Edits=8000, Edit Dataset=zsRE2026.0129.81
ICVPrompt=Zero-shot2025.0228.9
+ FVprompt=zero-shot2025.0228.9
PMETBackbone=LLaMA2, Number of Edits=8000, Edit Dataset=zsRE2026.0125.63
LoRAPrompt=Zero-shot2025.0224.8
FVPrompt=Zero-shot2025.0223
Transformerprompt=zero-shot2025.0223
TVPrompt=Zero-shot2025.0222
+ SVprompt=zero-shot2025.0222
SVPrompt=Zero-shot2025.0221.7
+ ICVprompt=zero-shot2025.0221.7
LTVPrompt=5-shot2025.0221.1
+ TVprompt=zero-shot2025.0221.1
+ LTVprompt=few-shot2025.0221.1
FVPrompt=5-shot2025.0220.3
SVPrompt=5-shot2025.0220.3
Transformerprompt=few-shot2025.0220.3
+ ICVprompt=few-shot2025.0220.3
MEMITBackbone=LLaMA2, Number of Edits=8000, Edit Dataset=zsRE2026.0120.23
ICVPrompt=5-shot2025.0220
+ FVprompt=few-shot2025.0220
MENDBackbone=LLaMA2, Number of Edits=8000, Edit Dataset=zsRE2026.0119.74
TVPrompt=5-shot2025.0219.5
+ SVprompt=few-shot2025.0219.5
LoRAPrompt=5-shot2025.0218.6
TransformerPrompt=5-shot2025.0218
Standard Promptingprompt=few-shot2025.0218
TransformerPrompt=Zero-shot2025.0210.2
Standard Promptingprompt=zero-shot2025.0210.2