添加千问新模型得分

README.md

@@ -166,6 +166,7 @@ $$
 
 | 模型 (Model) | 加权准确率 (Weighted Acc) | 备注 (Notes) |
 | :--- | :---: | :--- |
+| **Qwen3.5-35B-A3B** | 40.23% | Retries: 3 |
 | **Qwen3-VL-32B-Instruct** | 35.16% | Retries: 3 |
 | **GLM-4.5V-INT8** | 19.53% | Retries: 3 |
 

README_EN.md

@@ -145,6 +145,7 @@ $$
 
 | Model | Weighted Accuracy | Notes |
 | :--- | :---: | :--- |
+| **Qwen3.5-35B-A3B** | 40.23% | Retries: 3 |
 | **Qwen3-VL-32B-Instruct** | 35.16% | Retries: 3 |
 | **GLM-4.5V-INT8** | 19.53% | Retries: 3 |
 

run_eval.py

@@ -282,7 +282,7 @@ def main():
 
             for attempt in range(args.max_retries + 1):
                 attempts_used = attempt + 1
-                current_temp = 0.1 if attempt == 0 else 0.4
+                current_temp = 0.6 if attempt == 0 else 0.7
                 
                 # 构造本次请求的消息列表
                 messages = base_messages + retry_context