List your company

General Knowledge on MMLU (pass@1)

72.48pass@1SPINE
17.006431.408245.8160.2118Nov 22, 2025Dec 18, 2025Jan 13, 2026Feb 8, 2026Mar 6, 2026Apr 1, 2026Apr 27, 2026

Evaluation results

MethodLinks
SPINEBase Model=Qwen3-1.7B2025.1172.48
LMSIBase Model=Qwen3-1.7B2025.1171.74
TTRLBase Model=Qwen3-1.7B2025.1171.19
SEALONGBase Model=Qwen3-1.7B2025.1169.36
w/o MergingModels=Code2026.0268.39
Self-ConsistencyBase Model=Qwen3-1.7B2025.1165.39
Task ArithmeticModels=Fuse2026.0265.17
Dare Task ArithmeticModels=Fuse2026.0264.85
Dare Ties MergingModels=Fuse2026.0263.76
SCF-RKLModels=Fuse2026.0263.47
VSBToken budget=3, Number of shots=0, Backbone=LLaDA-8B2026.0463.11
AdaBlockToken budget=3, Number of shots=02026.0462.98
VSBToken budget=3, Number of shots=0, Backbone=LLaDA-1.52026.0462.94
Blockwise-SFTToken budget=3, Number of shots=02026.0462.81
D2FToken budget=3, Number of shots=02026.0462.71
DAEDALToken budget=3, Number of shots=02026.0462.58
LLaDA-1.5Token budget=3, Number of shots=02026.0462.52
LLaDA-8BToken budget=3, Number of shots=02026.0462.49
w/o MergingModels=Math2026.0261.83
dLLM-VarToken budget=3, Number of shots=02026.0461.71
Ties MergingModels=Fuse2026.0258.94
No adaptationBase Model=Qwen3-1.7B2025.1158.16
SCEModels=Fuse2026.0257.12
Mistral-7B-Instruct-v0.2Models=base02026.0243.23
Dare Ties MergingModels=Fuse2026.0241.78
SCF-RKLModels=Fuse2026.0238.79
Task ArithmeticModels=Fuse2026.0226.29
Dare Task ArithmeticModels=Fuse2026.0226.29
Ties MergingModels=Fuse2026.0219.37
SCEModels=Fuse2026.0219.28
MathCoder2-Mistral-7BModels=base12026.0219.14