List your company

Question Answering on CSQA (test)

87.2AccuracyDLR
18.14436.0725471.928Jul 18, 2023Jan 5, 2024Jun 25, 2024Dec 14, 2024Jun 4, 2025Nov 23, 2025May 14, 2026

Evaluation results

MethodLinks
DLRModel=Qwen3-8B2026.0587.2
SFTModel=Qwen3-8B2026.0585.6
DLRModel=Qwen3-4B2026.0584
DLR (C=1)Model=Qwen3-8B2026.0583.8
TokenAssortedModel=Qwen3-8B2026.0583.1
PauseTokenModel=Qwen3-8B2026.0582.9
SFTModel=Qwen3-4B2026.0582.5
UnrestrictedMoESeed Model=LLaMA-3.1-8B2026.0581.9
MetaMoESeed Model=LLaMA-3.1-8B2026.0581.33
Expert I (CSQA)Seed Model=LLaMA-3.1-8B2026.0581.24
BTMSeed Model=LLaMA-3.1-8B2026.0581.24
DLR (C=1)Model=Qwen3-4B2026.0581.1
DLRModel=Llama3.2-3B2026.0581
TokenAssortedModel=Qwen3-4B2026.0580.6
ModelSoupSeed Model=LLaMA-3.1-8B2026.0580.26
PauseTokenModel=Qwen3-4B2026.0580.2
SFTModel=Llama3.2-3B2026.0579.8
DLR (C=1)Model=Llama3.2-3B2026.0579.6
LLAMA 2Size=70B2023.0778.5
LLaMA-2Model Size=70B, Evaluation Protocol=Few-Shot2023.1078.5
DLRModel=Qwen3-1.7B2026.0578.4
DLR (C=1)Model=Qwen3-1.7B2026.0577.4
TokenAssortedModel=Llama3.2-3B2026.0577.3
PauseTokenModel=Llama3.2-3B2026.0577.1
PauseTokenModel=Qwen3-1.7B2026.0576.4
TokenAssortedModel=Qwen3-1.7B2026.0576.3
BTXSeed Model=LLaMA-3.1-8B2026.0575.76
UnrestrictedMoESeed Model=LLaMA-3.2-3B2026.0575.51
FlexOlmoSeed Model=LLaMA-3.1-8B2026.0575.18
LLaMA-2Model Size=70B, Evaluation Protocol=LAT2023.1075.1
Expert I (CSQA)Seed Model=LLaMA-3.2-3B2026.0574.94
MetaMoESeed Model=LLaMA-3.2-3B2026.0574.94
BTMSeed Model=LLaMA-3.2-3B2026.0574.61
LLAMA 2Size=34B2023.0774.3
SFTModel=Qwen3-1.7B2026.0574.3
LLAMA 1Size=65B2023.0774
ModelSoupSeed Model=LLaMA-3.2-3B2026.0573.71
Expert III (SocialIQA)Seed Model=LLaMA-3.1-8B2026.0573.63
FlexOlmoSeed Model=LLaMA-3.2-3B2026.0573.3
LLAMA 1Size=33B2023.0772.5
Expert II (CosmosQA)Seed Model=LLaMA-3.1-8B2026.0571.42
DLRModel=Llama3.2-1B2026.0571.4
FalconSize=40B2023.0770.4
BTXSeed Model=LLaMA-3.2-3B2026.0569.62
ZeroShotSeed Model=LLaMA-3.1-8B2026.0569.37
LLaMA-2Model Size=13B, Evaluation Protocol=LAT2023.1068.3
DLR (C=1)Model=Llama3.2-1B2026.0568.3
LLAMA 2Size=13B2023.0767.3
LLaMA-2Model Size=13B, Evaluation Protocol=Few-Shot2023.1067.3
DLRModel=Qwen3-0.6B2026.0566.4
TokenAssortedModel=Llama3.2-1B2026.0565.6
Expert III (SocialIQA)Seed Model=LLaMA-3.2-3B2026.0565.44
SFTModel=Llama3.2-1B2026.0565.2
PauseTokenModel=Llama3.2-1B2026.0564.9
PauseTokenModel=Qwen3-0.6B2026.0564.7
TokenAssortedModel=Qwen3-0.6B2026.0564.2
SFTModel=Qwen3-0.6B2026.0564.1
DLR (C=1)Model=Qwen3-0.6B2026.0564
Expert II (CosmosQA)Seed Model=LLaMA-3.2-3B2026.0563.06
LLaMA-2Model Size=7B, Evaluation Protocol=LAT2023.1062.6
ZeroShotSeed Model=LLaMA-3.2-3B2026.0562.49
LLAMA 1Size=13B2023.0762
MPTSize=30B2023.0758.2
LLAMA 2Size=7B2023.0757.8
LLaMA-2Model Size=7B, Evaluation Protocol=Few-Shot2023.1057.8
LLAMA 1Size=7B2023.0733.6
MPTSize=7B2023.0721.3
FalconSize=7B2023.0720.8