List your company
ResearchFormerly WizWand

Science Question Answering on OpenBookQA

94.6AccuracyEVOSELECT
23.04841.62460.278.776Apr 17, 2025Jun 23, 2025Aug 29, 2025Nov 5, 2025Jan 11, 2026Mar 19, 2026May 26, 2026
Updated 4mo ago

Evaluation results

MethodLinks
EVOSELECTIteration=Iter. 2, Selection Ratio=0.2, Model Scale=14b base model2026.0494.6———————
AttributionIteration=Iter. 1, Selection Ratio=0.2, Model Scale=14b base model2026.0494.4———————
RandomIteration=Iter. 2, Selection Ratio=0.2, Model Scale=14b base model2026.0494.4———————
TSDSIteration=Iter. 2, Selection Ratio=0.5, Model Scale=14b base model2026.0494.4———————
EVOSELECTIteration=Iter. 2, Selection Ratio=0.5, Model Scale=14b base model2026.0494.4———————
EVOSELECTIteration=Iter. 1, Selection Ratio=0.5, Model Scale=14b base model2026.0494.2———————
RandomIteration=Iter. 1, Selection Ratio=0.5, Model Scale=14b base model2026.0493.4———————
AttributionIteration=Iter. 2, Selection Ratio=0.5, Model Scale=14b base model2026.0493.4———————
DiversityIteration=Iter. 2, Selection Ratio=0.5, Model Scale=14b base model2026.0493.4———————
AttributionIteration=Iter. 1, Selection Ratio=0.5, Model Scale=14b base model2026.0493.2———————
DiversityIteration=Iter. 1, Selection Ratio=0.5, Model Scale=14b base model2026.0493.2———————
RandomIteration=Iter. 2, Selection Ratio=0.5, Model Scale=14b base model2026.0493.2———————
RandomIteration=Iter. 1, Selection Ratio=0.2, Model Scale=14b base model2026.0493———————
EVOSELECTIteration=Iter. 1, Selection Ratio=0.2, Model Scale=14b base model2026.0493———————
AttributionIteration=Iter. 2, Selection Ratio=0.2, Model Scale=14b base model2026.0493———————
AllIteration=Iter. 1, Model Scale=14b base model2026.0492.6———————
Attr-DivIteration=Iter. 1, Selection Ratio=0.2, Model Scale=14b base model2026.0492.6———————
TSDSIteration=Iter. 1, Selection Ratio=0.5, Model Scale=14b base model2026.0492.4———————
Attr-DivIteration=Iter. 2, Selection Ratio=0.5, Model Scale=14b base model2026.0492.4———————
TSDSIteration=Iter. 2, Selection Ratio=0.2, Model Scale=14b base model2026.0492.4———————
DiversityIteration=Iter. 1, Selection Ratio=0.2, Model Scale=14b base model2026.0492.2———————
Attr-DivIteration=Iter. 1, Selection Ratio=0.5, Model Scale=14b base model2026.0492.2———————
TSDSIteration=Iter. 1, Selection Ratio=0.2, Model Scale=14b base model2026.0492.2———————
Attr-DivIteration=Iter. 2, Selection Ratio=0.2, Model Scale=14b base model2026.0492.2———————
DiversityIteration=Iter. 2, Selection Ratio=0.2, Model Scale=14b base model2026.0491.8———————
AllIteration=Iter. 2, Model Scale=14b base model2026.0491.6———————
BaseModel Scale=14b base model2026.0491.4———————
STaR-GATE-DMode=Thinking Mode2025.0669.0320.85.216.645.44.734.355.7
SwiftMode=Thinking Mode2025.0668.1496.56.166.866.495.895.996.52
STaR-GATEMode=Thinking Mode2025.0665.4992.56.016.686.355.85.786.36
RefitMode=Thinking Mode2025.0665.0496.55.846.636.125.585.626.25
DPOMode=Thinking Mode2025.0661.9510.65.096.215.354.824.475.55
StepDPOMode=Thinking Mode2025.0661.0611.54.96.145.064.564.295.33
DEEPSEEK-V2-LITEPARAMS=16B, Variant=ORIGINAL2025.1145.2———————
LLAMA-4-SCOUTPARAMS=109B, Variant=ORIGINAL2025.1145.2———————
QWEN-3-30B MOEPARAMS=30B, Variant=ORIGINAL2025.1145———————
LLAMA-4-SCOUTPARAMS=109B, Variant=FCSD2025.1144.4———————
LLAMA-4-SCOUTPARAMS=109B, Variant=SFT2025.1143.8———————
MobileMoE-LActive Parameters=922M, Total Parameters=5.3B, Few-shot count=0-shot2026.0542.8———————
QWEN-3-30B MOEPARAMS=30B, Variant=FCSD2025.1142.8———————
DEEPSEEK-V2-LITEPARAMS=16B, Variant=FCSD2025.1141.8———————
QWEN-3-30B MOEPARAMS=30B, Variant=SFT2025.1141.4———————
DEEPSEEK-V2-LITEPARAMS=16B, Variant=SFT2025.1141.2———————
CLIMBSize=950M2025.0441.2———————
MobileMoE-MActive Parameters=528M, Total Parameters=2.8B, Few-shot count=0-shot2026.0539.4———————
SmolLMSize=360M2025.0437.6———————
Llama-3.2Size=1.2B2025.0437.2———————
CLIMBSize=350M2025.0436.6———————
BaseMode=Thinking Mode2025.0636.2874.35.996.776.155.435.956.31
TinyLlamaSize=1.1B2025.0436———————
AMD-OLMoSize=1.2B2025.0435.8———————
Qwen2.5Size=490M2025.0435.2———————
MobileMoE-SActive Parameters=272M, Total Parameters=1.3B, Few-shot count=0-shot2026.0534.6———————
X-GRAM-50%Backbone Scale=1.15B Scale, Cfg=4x, # Params=1.15B+7.47B, Evaluation Protocol=Zero/Few-shot2026.0433.4———————
X-GRAM-50%Backbone Scale=0.73B Scale, Cfg=4x, # Params=0.73B+4.67B, Evaluation Protocol=Zero/Few-shot2026.0432———————
X-GRAM-100%Backbone Scale=1.15B Scale, Cfg=2x, # Params=1.15B+7.47B, Evaluation Protocol=Zero/Few-shot2026.0431.6———————
X-GRAM-50%Backbone Scale=1.15B Scale, Cfg=2x, # Params=1.15B+3.73B, Evaluation Protocol=Zero/Few-shot2026.0431———————
X-GRAM-100%Backbone Scale=1.15B Scale, Cfg=1x, # Params=1.15B+3.73B, Evaluation Protocol=Zero/Few-shot2026.0430.8———————
X-GRAM-100%Backbone Scale=1.15B Scale, Cfg=4x, # Params=1.15B+14.94B, Evaluation Protocol=Zero/Few-shot2026.0430.8———————
MDMZero-shot=true2026.0530.8———————
X-GRAM-100%Backbone Scale=0.73B Scale, Cfg=2x, # Params=0.73B+4.67B, Evaluation Protocol=Zero/Few-shot2026.0430.4———————
MoRTBackbone Scale=1.15B Scale, Cfg=2x, # Params=1.15B+7.47B, Evaluation Protocol=Zero/Few-shot2026.0430.4———————
X-GRAM-100%Backbone Scale=0.73B Scale, Cfg=4x, # Params=0.73B+9.33B, Evaluation Protocol=Zero/Few-shot2026.0430.2———————
EngramBackbone Scale=1.15B Scale, Cfg=4x, # Params=1.15B+14.94B, Evaluation Protocol=Zero/Few-shot2026.0429.8———————
EngramBackbone Scale=0.73B Scale, Cfg=2x, # Params=0.73B+4.67B, Evaluation Protocol=Zero/Few-shot2026.0429.6———————
MoRTBackbone Scale=0.73B Scale, Cfg=4x, # Params=0.73B+9.33B, Evaluation Protocol=Zero/Few-shot2026.0429.6———————
BaselineBackbone Scale=1.15B Scale, Cfg=N/A, # Params=1.15B, Evaluation Protocol=Zero/Few-shot2026.0429.2———————
RetokenBackbone Scale=0.73B Scale, Cfg=2x, # Params=0.73B+4.67B, Evaluation Protocol=Zero/Few-shot2026.0429———————
EngramBackbone Scale=0.73B Scale, Cfg=4x, # Params=0.73B+9.33B, Evaluation Protocol=Zero/Few-shot2026.0429———————
X-GRAM-50%Backbone Scale=0.73B Scale, Cfg=1x, # Params=0.73B+1.17B, Evaluation Protocol=Zero/Few-shot2026.0428.8———————
MoRTBackbone Scale=0.73B Scale, Cfg=2x, # Params=0.73B+4.67B, Evaluation Protocol=Zero/Few-shot2026.0428.8———————
RetokenBackbone Scale=1.15B Scale, Cfg=1x, # Params=1.15B+3.73B, Evaluation Protocol=Zero/Few-shot2026.0428.6———————
MoRTBackbone Scale=1.15B Scale, Cfg=1x, # Params=1.15B+3.73B, Evaluation Protocol=Zero/Few-shot2026.0428.6———————
X-GRAM-50%Backbone Scale=1.15B Scale, Cfg=1x, # Params=1.15B+1.87B, Evaluation Protocol=Zero/Few-shot2026.0428.6———————
RetokenBackbone Scale=1.15B Scale, Cfg=2x, # Params=1.15B+7.47B, Evaluation Protocol=Zero/Few-shot2026.0428.6———————
RetokenBackbone Scale=1.15B Scale, Cfg=4x, # Params=1.15B+14.94B, Evaluation Protocol=Zero/Few-shot2026.0428.4———————
LoopMDMS=1, Zero-shot=true2026.0528.2———————
RetokenBackbone Scale=0.73B Scale, Cfg=1x, # Params=0.73B+2.33B, Evaluation Protocol=Zero/Few-shot2026.0428———————
MoRTBackbone Scale=0.73B Scale, Cfg=1x, # Params=0.73B+2.33B, Evaluation Protocol=Zero/Few-shot2026.0428———————
EngramBackbone Scale=1.15B Scale, Cfg=1x, # Params=1.15B+3.73B, Evaluation Protocol=Zero/Few-shot2026.0428———————
LoopMDMS=6, Zero-shot=true2026.0528———————
LoopMDMS=12, Zero-shot=true2026.0528———————
EngramBackbone Scale=0.73B Scale, Cfg=1x, # Params=0.73B+2.33B, Evaluation Protocol=Zero/Few-shot2026.0427.8———————
BaselineBackbone Scale=0.73B Scale, Cfg=N/A, # Params=0.73B, Evaluation Protocol=Zero/Few-shot2026.0427.6———————
X-GRAM-50%Backbone Scale=0.73B Scale, Cfg=2x, # Params=0.73B+2.33B, Evaluation Protocol=Zero/Few-shot2026.0427.4———————
EngramBackbone Scale=1.15B Scale, Cfg=2x, # Params=1.15B+7.47B, Evaluation Protocol=Zero/Few-shot2026.0427.4———————
MoRTBackbone Scale=1.15B Scale, Cfg=4x, # Params=1.15B+14.94B, Evaluation Protocol=Zero/Few-shot2026.0427.2———————
X-GRAM-100%Backbone Scale=0.73B Scale, Cfg=1x, # Params=0.73B+2.33B, Evaluation Protocol=Zero/Few-shot2026.0426.8———————
RetokenBackbone Scale=0.73B Scale, Cfg=4x, # Params=0.73B+9.33B, Evaluation Protocol=Zero/Few-shot2026.0425.8———————