Cultural Reasoning on FORK
82.61Mean@3x1-Qwen3-32B
Evaluation results
| Method | Links | |
|---|---|---|
| x1-Qwen3-32BReasoning Mode=Think, Task LoRA=+ Culture2026.04 | 82.61 | |
| Qwen3-32BReasoning Mode=Think2026.04 | 81.88 | |
| x1-Qwen3-32BReasoning Mode=Non-Think, Task LoRA=+ Culture2026.04 | 79.17 | |
| DeepSeek-V3.2Reasoning Mode=Think2026.04 | 78.8 | |
| Qwen3-32BReasoning Mode=Non-Think2026.04 | 78.26 | |
| x1-Qwen3-4BReasoning Mode=Think, Task LoRA=+ Culture2026.04 | 78.08 | |
| x1-Qwen3-14BReasoning Mode=Think, Task LoRA=+ Culture2026.04 | 76.81 | |
| x1-DeepSeek-R1-Distill-Llama-8BReasoning Mode=Think, Task LoRA=+ Culture2026.04 | 76.27 | |
| x1-Qwen3-4BReasoning Mode=Non-Think, Task LoRA=+ Culture2026.04 | 75.18 | |
| Qwen3-4BReasoning Mode=Non-Think2026.04 | 74.64 | |
| o4-mini-highReasoning Mode=Think2026.04 | 74.46 | |
| Qwen3-14BReasoning Mode=Non-Think2026.04 | 74.46 | |
| DeepSeek-R1-Distill-Llama-8BReasoning Mode=Think2026.04 | 74.09 | |
| Qwen3-14BReasoning Mode=Think2026.04 | 73.91 | |
| Qwen3-4BReasoning Mode=Think2026.04 | 73.73 | |
| x1-Qwen3-14BReasoning Mode=Non-Think, Task LoRA=+ Culture2026.04 | 73.12 | |
| DeepSeek-V3.2Reasoning Mode=Non-Think2026.04 | 72.83 | |
| x1-DeepSeek-R1-Distill-Qwen-7BReasoning Mode=Think, Task LoRA=+ Culture2026.04 | 58.7 | |
| x1-DeepSeek-R1-Distill-Llama-8BReasoning Mode=Non-Think, Task LoRA=+ Culture2026.04 | 57.07 | |
| DeepSeek-R1-Distill-Qwen-7BReasoning Mode=Think2026.04 | 55.25 | |
| x1-DeepSeek-R1-Distill-Qwen-7BReasoning Mode=Non-Think, Task LoRA=+ Culture2026.04 | 54.89 | |
| DeepSeek-R1-Distill-Qwen-7BReasoning Mode=Non-Think2026.04 | 40.04 | |
| DeepSeek-R1-Distill-Llama-8BReasoning Mode=Non-Think2026.04 | 37.86 |