Code Generation on CodeContest (Basic)
49.99Pass RateLogitsCoder
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LogitsCoderModel Category=Decoding-based Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 49.99 | 31.51 | |
| Contrastive DecodingModel Category=Decoding-based Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 45.12 | 28.83 | |
| LATSModel Category=Search-guided Reasoning Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 44.12 | 26.03 | |
| Guided DecodingModel Category=Decoding-based Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 43.6 | 26.03 | |
| self-playModel Category=Reflection-based Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 43.56 | 28.77 | |
| LDBModel Category=Reflection-based Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 41.39 | 27.4 | |
| MCTSModel Category=Search-guided Reasoning Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 41.14 | 23.29 | |
| RethinkMCTSModel Category=Search-guided Reasoning Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 40.03 | 23.81 | |
| Zero-shotModel Category=Base, Backbone=Qwen2.5-14B-Instruct, Evaluation Protocol=Zero-shot2026.02 | 34.38 | 19.18 | |
| ReflexionModel Category=Reflection-based Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 34.32 | 20.55 | |
| RAPModel Category=Reflection-based Models, Backbone=Qwen2.5-14B-Instruct, Rollout Budget=202026.02 | 29.54 | 16.44 |