Mathematical Reasoning on DART-Math DM-2
48Pass@k AccuracyPOLAR
Evaluation Results
| Method | Links | |
|---|---|---|
| POLARBackbone=LLaMA-3.2-3B-Instruct, k=52026.06 | 48 | |
| POLARp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 44 | |
| POLARp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 41.2 | |
| POLARp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 40.6 | |
| POLARp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 39 | |
| DR.LLMp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 36.8 | |
| POLARp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 34.6 | |
| POLARp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 34.6 | |
| POLARp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 31.4 | |
| DR.LLMp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 30.4 | |
| DR.LLMp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 30.2 | |
| Base (sampling)p@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 30.2 | |
| DR.LLMp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 29.8 | |
| Base (sampling)p@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 29.2 | |
| Base (tau=0)Backbone=LLaMA-3.2-3B-Instruct, k=12026.06 | 28.6 | |
| POLARp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 28 | |
| Base (sampling)p@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 26.8 | |
| POLARp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 26.2 | |
| Base (sampling)p@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 25.6 | |
| Base (sampling)p@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 25 | |
| Base (τ=0)p@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 24.6 | |
| Base (sampling)p@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 24.6 | |
| DR.LLMp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 24.6 | |
| POLARp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 24.4 | |
| Base (sampling)p@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 24.2 | |
| FlexiDepthp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 24 | |
| Base (sampling)p@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 22.8 | |
| FlexiDepthp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 22.4 | |
| Base (sampling)p@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 22 | |
| FlexiDepthp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 20 | |
| ShortGPTp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 18.6 | |
| ShortGPTp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 17.6 | |
| ShortGPTp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 16.6 | |
| Base (sampling)p@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 16 | |
| ShortGPTp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 15 | |
| FlexiDepthp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 15 | |
| DR.LLMp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 14.6 | |
| Base (τ=0)p@=1, Temperature=0, Backbone=Qwen2.5-3B-Instruct2026.06 | 13.6 | |
| DR.LLMp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 13 | |
| ShortGPTp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 12.2 | |
| DR.LLMp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 11.8 | |
| DR.LLMp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 9.8 | |
| FlexiDepthp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 8.8 | |
| DR.LLMp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 6.6 | |
| MindSkipp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 1.4 | |
| FlexiDepthp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 1 | |
| MindSkipp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 1 | |
| FlexiDepthp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0.8 | |
| FlexiDepthp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0.8 | |
| MindSkipp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0.8 | |
| MindSkipp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0.6 | |
| FlexiDepthp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0.4 | |
| MindSkipp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0.4 | |
| MindSkipp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0 | |
| MindSkipp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0 | |
| MindSkipp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0 | |
| MindSkipp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0 | |
| MindSkipp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0 | |
| FlexiDepthp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0 | |
| ShortGPTp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0 | |
| ShortGPTp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0 | |
| ShortGPTp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0 | |
| ShortGPTp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0 | |
| ShortGPTp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0 |