diff --git a/基于AI Agent开发范式的国产GPU大模型推理算子库优化/FlashInfer Benchmark实战：从性能基线到XPU-OJ冒烟提交.md b/基于AI Agent开发范式的国产GPU大模型推理算子库优化/FlashInfer Benchmark实战：从性能基线到XPU-OJ冒烟提交.md
index c993911..f9781a6 100644
--- a/基于AI Agent开发范式的国产GPU大模型推理算子库优化/FlashInfer Benchmark实战：从性能基线到XPU-OJ冒烟提交.md	
+++ b/基于AI Agent开发范式的国产GPU大模型推理算子库优化/FlashInfer Benchmark实战：从性能基线到XPU-OJ冒烟提交.md	
@@ -1082,7 +1082,7 @@ mv *.csv results/
 
 ### 参考 prompt
 
-[回退到 Step 8](#step%208提交%20oj%20冒烟代码)
+[*回退到 Step 8*](#step%208提交%20oj%20冒烟代码)
 
 ```plaintext
 # FlashInfer Ragged Prefill CUDA Kernel — Problem 20001