Mathematical Reasoning on DART-Math DM-4
40.4Pass@k AccuracyPOLAR
Evaluation Results
| Method | Links | |
|---|---|---|
| POLARBackbone=LLaMA-3.2-3B-Instruct, k=52026.06 | 40.4 | |
| DR.LLMBackbone=LLaMA-3.2-3B-Instruct, k=52026.06 | 33.4 | |
| POLARp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 25.4 | |
| POLARp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 23.6 | |
| DR.LLMp@=22.82026.06 | 22.8 | |
| POLARp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 22.4 | |
| POLARp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 20.8 | |
| DR.LLMp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 18.2 | |
| DR.LLMp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 18.2 | |
| DR.LLMp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 18.2 | |
| POLARp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 18 | |
| POLARp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 17.2 | |
| POLARp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 17.2 | |
| Base (sampling)p@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 15.8 | |
| Base (τ=0)p@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 15.2 | |
| DR.LLMp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 15.2 | |
| Base (sampling)p@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 15.2 | |
| Base (sampling)p@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 15 | |
| Base (sampling)p@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 14.8 | |
| POLARp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 14.8 | |
| Base (sampling)p@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 14 | |
| Base (sampling)p@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 13.8 | |
| Base (sampling)p@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 13.6 | |
| POLARp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 12.8 | |
| Base (sampling)p@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 12.6 | |
| Base (sampling)p@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 12.2 | |
| POLARp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 11.8 | |
| ShortGPTp@=5, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 10.2 | |
| ShortGPTp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 9.6 | |
| ShortGPTp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 8.6 | |
| Base (sampling)p@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 8.2 | |
| DR.LLMp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 8 | |
| Base (τ=0)p@=1, Temperature=0, Backbone=Qwen2.5-3B-Instruct2026.06 | 7.8 | |
| ShortGPTp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 7.2 | |
| DR.LLMp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 7 | |
| ShortGPTp@=1, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 5.4 | |
| DR.LLMp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 5.4 | |
| FlexiDepthp@=5, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 3.8 | |
| DR.LLMp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 3.6 | |
| FlexiDepthp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 3.4 | |
| FlexiDepthp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 2.8 | |
| FlexiDepthp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 2.2 | |
| DR.LLMp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 2.2 | |
| FlexiDepthp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 1 | |
| MindSkipp@=1, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0.4 | |
| MindSkipp@=2, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0.4 | |
| MindSkipp@=3, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0.4 | |
| MindSkipp@=4, Temperature=sampling, Backbone=Qwen2.5-3B-Instruct2026.06 | 0.4 | |
| MindSkipp@=5, 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 | |
| FlexiDepthp@=2, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0 | |
| FlexiDepthp@=3, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0 | |
| FlexiDepthp@=4, Backbone=Qwen1.5-MoE-A2.7B-Chat2026.06 | 0 | |
| FlexiDepthp@=5, 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 |