ShuaiBai623 commited on
Commit
ec57a51
·
verified ·
1 Parent(s): 32881f5

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -568,6 +568,7 @@ For more details, please refer to our blog post [Qwen3.5](https://qwen.ai/blog?i
568
  </table>
569
 
570
  <p style="margin-top:12px;font-size:11px;opacity:0.7">
 
571
  * MathVision: our model’s score is evaluated using a fixed prompt, e.g., “Please reason step by step, and put your final answer within \boxed{}.” For other models, we report the higher score between runs with and without the \boxed{} formatting.<br>
572
  * Experimental settings: For the Video benchmarks, we used top_p=0.95, top_k=20, presence_penalty=1.5, and temperature=1.0. All other benchmarks adopted the same hyperparameter configuration but with temperature=0.6 under the thinking mode. Under the no-thinking mode, the inference hyperparameters were set to top_p=0.8, top_k=20, presence_penalty=1.5, and temperature=0.7.<br>
573
  * Empty cells (--) indicate scores not yet available or not applicable.
 
568
  </table>
569
 
570
  <p style="margin-top:12px;font-size:11px;opacity:0.7">
571
+ * Dual scores: Thinking / No-thinking. Single scores: Thinking (except ScreenSpot Pro: No-thinking).<br>
572
  * MathVision: our model’s score is evaluated using a fixed prompt, e.g., “Please reason step by step, and put your final answer within \boxed{}.” For other models, we report the higher score between runs with and without the \boxed{} formatting.<br>
573
  * Experimental settings: For the Video benchmarks, we used top_p=0.95, top_k=20, presence_penalty=1.5, and temperature=1.0. All other benchmarks adopted the same hyperparameter configuration but with temperature=0.6 under the thinking mode. Under the no-thinking mode, the inference hyperparameters were set to top_p=0.8, top_k=20, presence_penalty=1.5, and temperature=0.7.<br>
574
  * Empty cells (--) indicate scores not yet available or not applicable.