diff --git a/example/001-example/reduce_sum_performance.yaml b/example/001-example/reduce_sum_performance.yaml
deleted file mode 100644
index efa6879..0000000
--- a/example/001-example/reduce_sum_performance.yaml
+++ /dev/null
@@ -1,26 +0,0 @@
-# ReduceSum算法性能测试结果
-# 生成时间: 2025-08-19 15:42:19
-
-algorithm: "ReduceSum"
-data_types:
-  input: "float"
-  output: "float"
-formulas:
-  throughput: "elements / time(s) / 1e9 (G/s)"
-performance_data:
-  - data_size: 1000000
-    time_ms: 0.059418
-    throughput_gps: 16.830029
-    data_type: "float"
-  - data_size: 134217728
-    time_ms: 0.884915
-    throughput_gps: 151.672977
-    data_type: "float"
-  - data_size: 536870912
-    time_ms: 2.811878
-    throughput_gps: 190.929630
-    data_type: "float"
-  - data_size: 1073741824
-    time_ms: 5.376102
-    throughput_gps: 199.724956
-    data_type: "float"
diff --git a/example/001-example/sort_pair_performance.yaml b/example/001-example/sort_pair_performance.yaml
deleted file mode 100644
index 279fb3e..0000000
--- a/example/001-example/sort_pair_performance.yaml
+++ /dev/null
@@ -1,46 +0,0 @@
-# SortPair算法性能测试结果
-# 生成时间: 2025-08-19 15:44:51
-
-algorithm: "SortPair"
-data_types:
-  key_type: "float"
-  value_type: "uint32_t"
-formulas:
-  throughput: "elements / time(s) / 1e9 (G/s)"
-performance_data:
-  - data_size: 1000000
-    ascending:
-      time_ms: 0.645709
-      throughput_gps: 1.548686
-    descending:
-      time_ms: 0.643405
-      throughput_gps: 1.554231
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 134217728
-    ascending:
-      time_ms: 48.160717
-      throughput_gps: 2.786871
-    descending:
-      time_ms: 48.198193
-      throughput_gps: 2.784705
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 536870912
-    ascending:
-      time_ms: 192.608109
-      throughput_gps: 2.787374
-    descending:
-      time_ms: 192.794830
-      throughput_gps: 2.784675
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1073741824
-    ascending:
-      time_ms: 391.377167
-      throughput_gps: 2.743496
-    descending:
-      time_ms: 391.900879
-      throughput_gps: 2.739830
-    key_type: "float"
-    value_type: "uint32_t"
diff --git a/example/001-example/topk_pair_performance.yaml b/example/001-example/topk_pair_performance.yaml
deleted file mode 100644
index c9540eb..0000000
--- a/example/001-example/topk_pair_performance.yaml
+++ /dev/null
@@ -1,210 +0,0 @@
-# TopkPair算法性能测试结果
-# 生成时间: 2025-08-19 15:48:37
-
-algorithm: "TopkPair"
-data_types:
-  key_type: "float"
-  value_type: "uint32_t"
-formulas:
-  throughput: "elements / time(s) / 1e9 (G/s)"
-performance_data:
-  - data_size: 1000000
-    k_value: 32
-    ascending:
-      time_ms: 0.678861
-      throughput_gps: 1.473056
-    descending:
-      time_ms: 0.686054
-      throughput_gps: 1.457610
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1000000
-    k_value: 50
-    ascending:
-      time_ms: 0.707123
-      throughput_gps: 1.414181
-    descending:
-      time_ms: 0.679578
-      throughput_gps: 1.471502
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1000000
-    k_value: 100
-    ascending:
-      time_ms: 0.700877
-      throughput_gps: 1.426784
-    descending:
-      time_ms: 0.671693
-      throughput_gps: 1.488776
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1000000
-    k_value: 256
-    ascending:
-      time_ms: 0.706432
-      throughput_gps: 1.415564
-    descending:
-      time_ms: 0.689306
-      throughput_gps: 1.450735
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1000000
-    k_value: 1024
-    ascending:
-      time_ms: 0.663194
-      throughput_gps: 1.507855
-    descending:
-      time_ms: 0.660531
-      throughput_gps: 1.513933
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 134217728
-    k_value: 32
-    ascending:
-      time_ms: 48.147915
-      throughput_gps: 2.787612
-    descending:
-      time_ms: 48.289310
-      throughput_gps: 2.779450
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 134217728
-    k_value: 50
-    ascending:
-      time_ms: 48.196999
-      throughput_gps: 2.784774
-    descending:
-      time_ms: 48.226074
-      throughput_gps: 2.783095
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 134217728
-    k_value: 100
-    ascending:
-      time_ms: 48.210712
-      throughput_gps: 2.783981
-    descending:
-      time_ms: 48.247169
-      throughput_gps: 2.781878
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 134217728
-    k_value: 256
-    ascending:
-      time_ms: 48.231018
-      throughput_gps: 2.782809
-    descending:
-      time_ms: 48.235191
-      throughput_gps: 2.782569
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 134217728
-    k_value: 1024
-    ascending:
-      time_ms: 48.243660
-      throughput_gps: 2.782080
-    descending:
-      time_ms: 48.287209
-      throughput_gps: 2.779571
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 536870912
-    k_value: 32
-    ascending:
-      time_ms: 192.694534
-      throughput_gps: 2.786124
-    descending:
-      time_ms: 192.709503
-      throughput_gps: 2.785908
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 536870912
-    k_value: 50
-    ascending:
-      time_ms: 192.567810
-      throughput_gps: 2.787958
-    descending:
-      time_ms: 192.672104
-      throughput_gps: 2.786449
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 536870912
-    k_value: 100
-    ascending:
-      time_ms: 192.633591
-      throughput_gps: 2.787006
-    descending:
-      time_ms: 192.675278
-      throughput_gps: 2.786403
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 536870912
-    k_value: 256
-    ascending:
-      time_ms: 192.612732
-      throughput_gps: 2.787307
-    descending:
-      time_ms: 192.753647
-      throughput_gps: 2.785270
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 536870912
-    k_value: 1024
-    ascending:
-      time_ms: 192.627899
-      throughput_gps: 2.787088
-    descending:
-      time_ms: 192.446716
-      throughput_gps: 2.789712
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1073741824
-    k_value: 32
-    ascending:
-      time_ms: 392.105988
-      throughput_gps: 2.738397
-    descending:
-      time_ms: 392.092041
-      throughput_gps: 2.738494
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1073741824
-    k_value: 50
-    ascending:
-      time_ms: 391.682678
-      throughput_gps: 2.741356
-    descending:
-      time_ms: 392.191681
-      throughput_gps: 2.737799
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1073741824
-    k_value: 100
-    ascending:
-      time_ms: 391.761108
-      throughput_gps: 2.740808
-    descending:
-      time_ms: 392.304626
-      throughput_gps: 2.737010
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1073741824
-    k_value: 256
-    ascending:
-      time_ms: 391.693726
-      throughput_gps: 2.741279
-    descending:
-      time_ms: 392.074707
-      throughput_gps: 2.738615
-    key_type: "float"
-    value_type: "uint32_t"
-  - data_size: 1073741824
-    k_value: 1024
-    ascending:
-      time_ms: 391.690765
-      throughput_gps: 2.741300
-    descending:
-      time_ms: 392.122711
-      throughput_gps: 2.738280
-    key_type: "float"
-    value_type: "uint32_t"
diff --git a/example/a guide to GPUKernelContest/README.md b/example/README.md
similarity index 100%
rename from example/a guide to GPUKernelContest/README.md
rename to example/README.md
diff --git a/example/a guide to GPUKernelContest/build_and_run.sh b/example/build_and_run.sh
similarity index 100%
rename from example/a guide to GPUKernelContest/build_and_run.sh
rename to example/build_and_run.sh
diff --git a/example/a guide to GPUKernelContest/build_and_run_reduce_sum.sh b/example/build_and_run_reduce_sum.sh
similarity index 100%
rename from example/a guide to GPUKernelContest/build_and_run_reduce_sum.sh
rename to example/build_and_run_reduce_sum.sh
diff --git a/example/a guide to GPUKernelContest/build_and_run_sort_pair.sh b/example/build_and_run_sort_pair.sh
similarity index 100%
rename from example/a guide to GPUKernelContest/build_and_run_sort_pair.sh
rename to example/build_and_run_sort_pair.sh
diff --git a/example/a guide to GPUKernelContest/build_and_run_topk_pair.sh b/example/build_and_run_topk_pair.sh
similarity index 100%
rename from example/a guide to GPUKernelContest/build_and_run_topk_pair.sh
rename to example/build_and_run_topk_pair.sh
diff --git a/example/a guide to GPUKernelContest/build_common.sh b/example/build_common.sh
similarity index 97%
rename from example/a guide to GPUKernelContest/build_common.sh
rename to example/build_common.sh
index 1564496..c189158 100644
--- a/example/a guide to GPUKernelContest/build_common.sh	
+++ b/example/build_common.sh
@@ -33,7 +33,7 @@ COMPILER=${COMPILER:-mxcc}
 #COMPILER_FLAGS=${COMPILER_FLAGS:--O3 -std=c++17 --extended-lambda} # not run all test for easy debug
 COMPILER_FLAGS=${COMPILER_FLAGS:--O3 -std=c++17 --extended-lambda -DRUN_FULL_TEST} 
 
-INCLUDE_DIR=${INCLUDE_DIR:-src}
+INCLUDE_DIR=${INCLUDE_DIR:-}
 BUILD_DIR=${BUILD_DIR:-build}
 
 # 编译单个算法的通用函数
@@ -49,7 +49,7 @@ compile_algorithm() {
     mkdir -p "$BUILD_DIR"
     
     # 编译命令
-    local compile_cmd="$COMPILER $COMPILER_FLAGS -I$INCLUDE_DIR src/$source_file -o $target_file"
+    local compile_cmd="$COMPILER $COMPILER_FLAGS -I$INCLUDE_DIR $source_file -o $target_file"
     
     print_info "执行: $compile_cmd"
     
diff --git a/example/a guide to GPUKernelContest/competition_parallel_algorithms.md b/example/competition_parallel_algorithms.md
similarity index 100%
rename from example/a guide to GPUKernelContest/competition_parallel_algorithms.md
rename to example/competition_parallel_algorithms.md
diff --git a/example/a guide to GPUKernelContest/src/performance_utils.h b/example/performance_utils.h
similarity index 100%
rename from example/a guide to GPUKernelContest/src/performance_utils.h
rename to example/performance_utils.h
diff --git a/example/a guide to GPUKernelContest/src/reduce_sum_algorithm.maca b/example/reduce_sum_algorithm.maca
similarity index 100%
rename from example/a guide to GPUKernelContest/src/reduce_sum_algorithm.maca
rename to example/reduce_sum_algorithm.maca
diff --git a/example/a guide to GPUKernelContest/src/sort_pair_algorithm.maca b/example/sort_pair_algorithm.maca
similarity index 100%
rename from example/a guide to GPUKernelContest/src/sort_pair_algorithm.maca
rename to example/sort_pair_algorithm.maca
diff --git a/example/a guide to GPUKernelContest/src/test_utils.h b/example/test_utils.h
similarity index 100%
rename from example/a guide to GPUKernelContest/src/test_utils.h
rename to example/test_utils.h
diff --git a/example/a guide to GPUKernelContest/src/topk_pair_algorithm.maca b/example/topk_pair_algorithm.maca
similarity index 100%
rename from example/a guide to GPUKernelContest/src/topk_pair_algorithm.maca
rename to example/topk_pair_algorithm.maca
diff --git a/example/a guide to GPUKernelContest/src/yaml_reporter.h b/example/yaml_reporter.h
similarity index 100%
rename from example/a guide to GPUKernelContest/src/yaml_reporter.h
rename to example/yaml_reporter.h