List your company

Abstract Reasoning on AbsR

83.41AbsR AccuracyKALE
27.343641.899356.45571.0107Mar 14, 2024Jul 3, 2024Oct 23, 2024Feb 11, 2025Jun 3, 2025Sep 22, 2025Jan 12, 2026

Evaluation results

MethodLinks
KALEBackbone=Orca2 7B, Category=Augmented-based2026.0183.41—
KALEBackbone=OLMOE 7B, Category=Augmented-based2026.0181.99—
KALEBackbone=Gemma2 9B, Category=Augmented-based2026.0181.52—
KG-SFTBackbone=Orca2 7B, Category=SFT-based2026.0178.91—
DMTBackbone=Gemma2 9B, Category=SFT-based2026.0177.13—
GPT-3.5Size=>20B2024.0377.0884
MeanLearnBackbone=Orca2 7B, Category=SFT-based2026.0177.01—
STaRBackbone=Gemma2 9B, Category=Augmented-based2026.0176.66—
SDFTBackbone=Orca2 7B, Category=SFT-based2026.0176.66—
SDFTBackbone=Gemma2 9B, Category=SFT-based2026.0175.83—
DMTBackbone=Orca2 7B, Category=SFT-based2026.0175.24—
KG-SFTBackbone=Gemma2 9B, Category=SFT-based2026.0174.76—
MeanLearnSize=8B2024.0373.9684.5
LLaMA-3Size=8B2024.0372.9282.5
TOGBackbone=Gemma2 9B, Category=Retrieval-based2026.0172.04—
MeanLearnBackbone=Gemma2 9B, Category=SFT-based2026.0172.04—
STaRBackbone=Orca2 7B, Category=Augmented-based2026.0171.68—
MeanLearnBackbone=OLMOE 7B, Category=SFT-based2026.0171.09—
GPT3MixBackbone=Orca2 7B, Category=Augmented-based2026.0169.79—
CoTBackbone=Orca2 7B, Category=Prompt-based2026.0167.77—
CoTBackbone=Gemma2 9B, Category=Prompt-based2026.0167.54—
GraphRAGBackbone=Orca2 7B, Category=Retrieval-based2026.0167.06—
StructGPTBackbone=Orca2 7B, Category=Retrieval-based2026.0165.17—
PaLM-2Size=>20B2024.0364.5875.76
MeanLearnSize=7B2024.0364.5877
GraphRAGBackbone=Gemma2 9B, Category=Retrieval-based2026.0164.57—
SFTBackbone=Orca2 7B, Category=SFT-based2026.0163.98—
GPT3MixBackbone=OLMOE 7B, Category=Augmented-based2026.0162.2—
AugGPTBackbone=OLMOE 7B, Category=Augmented-based2026.0161.73—
AugGPTBackbone=Orca2 7B, Category=Augmented-based2026.0161.73—
LLaMA-2Size=13B2024.0361.4673.5
VicunaSize=13B2024.0361.4674
SFTBackbone=Gemma2 9B, Category=SFT-based2026.0161.37—
VanillaBackbone=Orca2 7B, Category=Prompt-based2026.0161.37—
KG-SFTBackbone=OLMOE 7B, Category=SFT-based2026.0161.26—
DMTBackbone=OLMOE 7B, Category=SFT-based2026.0160.43—
Orca-2Size=7B2024.0360.4273.5
SDFTBackbone=OLMOE 7B, Category=SFT-based2026.0159.95—
GPT3MixBackbone=Gemma2 9B, Category=Augmented-based2026.0159.72—
AugGPTBackbone=Gemma2 9B, Category=Augmented-based2026.0159.6—
TOGBackbone=Orca2 7B, Category=Retrieval-based2026.0159.6—
WizardLMSize=13B2024.0359.3874
StructGPTBackbone=Gemma2 9B, Category=Retrieval-based2026.0159.24—
STaRBackbone=OLMOE 7B, Category=Augmented-based2026.0159.24—
VicunaSize=7B2024.0358.3375
GraphRAGBackbone=OLMOE 7B, Category=Retrieval-based2026.0157.82—
StructGPTBackbone=OLMOE 7B, Category=Retrieval-based2026.0156.87—
TOGBackbone=OLMOE 7B, Category=Retrieval-based2026.0154.5—
SFTBackbone=OLMOE 7B, Category=SFT-based2026.0153.31—
WizardLMSize=7B2024.0353.2465
VanillaBackbone=Gemma2 9B, Category=Prompt-based2026.0152.49—
CoTBackbone=OLMOE 7B, Category=Prompt-based2026.0151.06—
VanillaBackbone=OLMOE 7B, Category=Prompt-based2026.0149.88—
MeanLearnSize=13B2024.0345.8267.17
LLaMA-2Size=7B2024.0335.4250
Orca-2Size=13B2024.0329.566.5
COT-DistillBackbone=Llama3 8B2026.01—80.64
Distill-SBSBackbone=Llama3 8B2026.01—77.61
KALEBackbone=Llama3 8B2026.01—83.62
KALEBackbone=Qwen3-32B-thinking2026.01—93.6
PDBackbone=Llama3 8B2026.01—75.31
SFTBackbone=Qwen3-32B-thinking2026.01—85.82
VanillaBackbone=Qwen3-32B-thinking2026.01—82.33