List your company

Multistep Soft Reasoning on MUSR (Accuracy %)

50.77Accuracy (Multi-choice)TMAP
36.761240.398144.03547.6719May 10, 2026

Evaluation results

MethodLinks
TMAPBackbone=Mistral-v0.2-7B2026.0550.77—
Data-WhispererBackbone=Mistral-v0.2-7B2026.0547.89—
BaseBackbone=Mistral-v0.2-7B2026.0547.49—
AlpagasusBackbone=Mistral-v0.2-7B2026.0546.43—
Full (100%)Backbone=Mistral-v0.2-7B2026.0546.3—
RandomBackbone=Mistral-v0.2-7B2026.0546.16—
Full (100%)Backbone=Qwen2.5-7B2026.0546.03—
LongestBackbone=Mistral-v0.2-7B2026.0545.9—
TMAPBackbone=Qwen2.5-7B2026.0545.9—
DEITABackbone=Mistral-v0.2-7B2026.0545.44—
IFDBackbone=Qwen2.5-7B2026.0544.84—
Data-WhispererBackbone=Qwen2.5-7B2026.0544.56—
IFDBackbone=Mistral-v0.2-7B2026.0544.44—
AlpagasusBackbone=Qwen2.5-7B2026.0544.02—
LongestBackbone=Qwen2.5-7B2026.0543.92—
RandomBackbone=Qwen2.5-7B2026.0543.77—
BaseBackbone=Qwen2.5-7B2026.0543.67—
DEITABackbone=Qwen2.5-7B2026.0543.52—
TMAPBackbone=Llama3-8B2026.0542.99—
AlpagasusBackbone=Llama3-8B2026.0542.86—
IFDBackbone=Llama3-8B2026.0542.72—
Data-WhispererBackbone=Llama3-8B2026.0541.8—
RandomBackbone=Llama3-8B2026.0541.67—
Full (100%)Backbone=Llama3-8B2026.0540.21—
DEITABackbone=Llama3-8B2026.0538.89—
BaseBackbone=Llama3-8B2026.0537.96—
LongestBackbone=Llama3-8B2026.0537.3—
CoSpaDiBase Model=Qwen 3 8B, Compression Ratio=0.22025.09—42.1
CoSpaDiBase Model=Qwen 3 8B, Compression Ratio=0.32025.09—38.4
CoSpaDiBase Model=Qwen 3 8B, Compression Ratio=0.42025.09—38.1
CoSpaDiBase Model=Qwen 3 14B, Compression Ratio=0.22025.09—40.7
CoSpaDiBase Model=Qwen 3 14B, Compression Ratio=0.42025.09—38
Qwen 3 14BCompression Ratio=Full2025.09—40.7
Qwen 3 8BCompression Ratio=Full2025.09—43.1
SVDLLMBase Model=Qwen 3 8B, Compression Ratio=0.22025.09—39.8
SVDLLMBase Model=Qwen 3 8B, Compression Ratio=0.32025.09—41.4
SVDLLMBase Model=Qwen 3 8B, Compression Ratio=0.42025.09—37.7
SVDLLMBase Model=Qwen 3 14B, Compression Ratio=0.22025.09—41.5
SVDLLMBase Model=Qwen 3 14B, Compression Ratio=0.42025.09—39.7