List your company

Mathematical Reasoning on DART-Math DM-5

45.8Pass@k AccuracyPOLAR
-1.83210.53422.935.266Jun 4, 2026

Evaluation results

MethodLinks
POLARBackbone=LLaMA-3.2-3B-Instruct, k=52026.0645.8
POLARp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0623.2
POLARp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.0622.8
POLARp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0622.6
DR.LLMp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0621.6
POLARp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0621.6
POLARp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.0621
DR.LLMp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0620
POLARp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0619.8
POLARp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.0619
DR.LLMp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0618.8
DR.LLMp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0618
POLARp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.0616.4
POLARp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0615.4
Base (τ=0)p@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0614.2
DR.LLMp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0614.2
Base (sampling)p@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.0613
POLARp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.0612.8
Base (sampling)p@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0611.8
Base (sampling)p@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.0611.2
Base (sampling)p@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.0610.6
DR.LLMp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.0610.2
Base (sampling)p@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.0610
Base (sampling)p@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.069.6
DR.LLMp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.069.4
Base (sampling)p@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.068.6
Base (sampling)p@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.068.2
ShortGPTp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.067.6
DR.LLMp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.067.4
DR.LLMp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.067.2
ShortGPTp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.067
Base (sampling)p@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.066.6
ShortGPTp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.066.2
Base (sampling)p@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.065.4
ShortGPTp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.065.2
Base (τ=0)p@=1, Temperature=0, Backbone=Qwen2.5-3B-Instruct2026.065.2
FlexiDepthp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.065
FlexiDepthp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.064.6
DR.LLMp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.064.2
ShortGPTBackbone=LLaMA-3.2-3B-Instruct, k=12026.063.8
ShortGPTp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.063.4
FlexiDepthp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.063
FlexiDepthp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.061.8
FlexiDepthp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.061.2
MindSkipp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
MindSkipp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
MindSkipp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
MindSkipp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
MindSkipp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
FlexiDepthp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
FlexiDepthp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
FlexiDepthp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
FlexiDepthp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
FlexiDepthp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.060
ShortGPTp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060
ShortGPTp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060
ShortGPTp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060
ShortGPTp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060
ShortGPTp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060
MindSkipp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060
MindSkipp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060
MindSkipp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060
MindSkipp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060
MindSkipp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.060