List your company

Reasoning on MuSR

71.89AccuracyLLaDA2.1-mini
30.903641.544352.18562.8257Nov 14, 2025Dec 23, 2025Jan 31, 2026Mar 11, 2026Apr 19, 2026May 28, 2026Jul 6, 2026

Evaluation results

MethodLinks
LLaDA2.1-minimode=Q Mode2026.0271.891.56
LLaDA2.0-mini2026.0271.481.45
Ling-mini-2.02026.0271.36—
LLaDA2.1-minimode=S Mode2026.0270.432.48
Qwen3-8Bno think=true2026.0270.02—
Baseline with CoTBackbone=Llama 3 70B2026.0766.8—
MALLM (Judge)Backbone=Llama 3 70B, Discussion Paradigm=Memory, Decision Protocol=Judge2026.0764.2—
BaselineBackbone=Llama 3 70B2026.0763.2—
MALLM (Majority Consensus)Backbone=Llama 3 70B, Discussion Paradigm=Memory, Decision Protocol=Majority Consensus2026.0761.3—
MALLM (Unanimity Consensus)Backbone=Llama 3 70B, Discussion Paradigm=Memory, Decision Protocol=Unanimity Consensus2026.0761.3—
MALLM (Average Consensus)Backbone=Llama 3 70B, Discussion Paradigm=Memory2026.0760.9—
MALLM (Supermajority Consensus)Backbone=Llama 3 70B, Discussion Paradigm=Memory, Decision Protocol=Supermajority Consensus2026.0760.2—
MALLM (Cumulative Voting)Backbone=Llama 3 70B, Discussion Paradigm=Memory, Decision Protocol=Cumulative Voting2026.0760—
MALLM (Average Voting)Backbone=Llama 3 70B, Discussion Paradigm=Memory2026.0759.6—
MALLM (Ranked Voting)Backbone=Llama 3 70B, Discussion Paradigm=Memory, Decision Protocol=Ranked Voting2026.0759.5—
MALLM (Simple Voting)Backbone=Llama 3 70B, Discussion Paradigm=Memory, Decision Protocol=Simple Voting2026.0759.3—
Qwen3-4BLoss function=base+aux2026.0255.46—
Qwen3-4BLoss function=base2026.0254.14—
Qwen3-8BLoss function=base+aux2026.0253.5—
Qwen3-8BLoss function=base2026.0249.78—
MALLM (Approval Voting)Backbone=Llama 3 70B, Discussion Paradigm=Memory, Decision Protocol=Approval Voting2026.0749.1—
LLaMA3.1-8BLoss function=base+aux2026.0246.17—
LLaMA3.1-8BLoss function=base2026.0244.87—
UM-190kBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1136.28—
UM-187kBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1136.07—
UltraFBBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1135.94—
UM-170kBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1135.94—
SFTBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1135.85—
TuluDPOBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1135.07—
ORPOBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1134.98—
HelpSteerBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1134.47—
CodePrefBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1132.48—