Hugging Face's logo

Datasets:
AINativeOps
/
AINativeBench

Dataset card Files
xet
 Community
王子睿 commited on
Commit
0d4cded
·
1 Parent(s): 165e0b7

restructure + add files

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. data/processed/RQ1/EmailResponder-MCP/evaluate_success.py +1 -1
  2. data/processed/RQ1/EmailResponder-MCP/{success-finish_rate.csv → success_rate.csv} +0 -0
  3. data/processed/RQ1/EmailResponder/evaluate_success.py +1 -1
  4. data/processed/RQ1/EmailResponder/{success-finish_rate.csv → success_rate.csv} +0 -0
  5. data/processed/RQ1/GameBuilder-MCP/evaluate_success.py +1 -1
  6. data/processed/RQ1/GameBuilder-MCP/{success-finish_rate.csv → success_rate.csv} +0 -0
  7. data/processed/RQ1/GameBuilder/evaluate_success.py +1 -1
  8. data/processed/RQ1/GameBuilder/{success-finish_rate.csv → success_rate.csv} +0 -0
  9. data/processed/RQ1/MarkdownValidator-MCP/evaluate_success.py +1 -1
  10. data/processed/RQ1/MarkdownValidator-MCP/{success-finish_rate.csv → success_rate.csv} +0 -0
  11. data/processed/RQ1/MarkdownValidator/evaluate_success.py +1 -1
  12. data/processed/RQ1/MarkdownValidator/{success-finish_rate.csv → success_rate.csv} +0 -0
  13. data/processed/RQ1/RadarCharts/All_Open_vs_Closed_summaries.md +34 -34
  14. data/processed/RQ1/RadarCharts/All_Series_Average_tables.md +60 -42
  15. data/processed/RQ1/RadarCharts/BookWriter-A2A_radar.pdf +2 -2
  16. data/processed/RQ1/RadarCharts/BookWriter-H_A2A_radar.pdf +2 -2
  17. data/processed/RQ1/RadarCharts/BookWriter-MCP_radar.pdf +2 -2
  18. data/processed/RQ1/RadarCharts/BookWriter_Series_Average_radar.pdf +1 -1
  19. data/processed/RQ1/RadarCharts/Comparison_A2A_vs_A2A_mix.md +0 -68
  20. data/processed/RQ1/RadarCharts/Comparison_A2A_vs_H_A2A.md +539 -0
  21. data/processed/RQ1/RadarCharts/Comparison_MCP_vs_A2A.md +482 -11
  22. data/processed/RQ1/RadarCharts/Comparison_MCP_vs_Hardcoded.md +109 -109
  23. data/processed/RQ1/RadarCharts/Comparison_Powerful_vs_Lightweight.md +32 -60
  24. data/processed/RQ1/RadarCharts/Comparison_Reasoning_vs_NonReasoning.md +62 -118
  25. data/processed/RQ1/RadarCharts/EmailResponder-MCP_radar.pdf +2 -2
  26. data/processed/RQ1/RadarCharts/EmailResponder_Series_Average_radar.pdf +2 -2
  27. data/processed/RQ1/RadarCharts/EmailResponder_radar.pdf +2 -2
  28. data/processed/RQ1/RadarCharts/GameBuilder-MCP_radar.pdf +2 -2
  29. data/processed/RQ1/RadarCharts/GameBuilder_Series_Average_radar.pdf +2 -2
  30. data/processed/RQ1/RadarCharts/GameBuilder_radar.pdf +2 -2
  31. data/processed/RQ1/RadarCharts/LandingPageGenerator-A2A_radar.pdf +2 -2
  32. data/processed/RQ1/RadarCharts/LandingPageGenerator-H_A2A_radar.pdf +2 -2
  33. data/processed/RQ1/RadarCharts/LandingPageGenerator-MCP_radar.pdf +2 -2
  34. data/processed/RQ1/RadarCharts/LandingPageGenerator_Series_Average_radar.pdf +2 -2
  35. data/processed/RQ1/RadarCharts/MarkdownValidator-MCP_radar.pdf +2 -2
  36. data/processed/RQ1/RadarCharts/MarkdownValidator_Series_Average_radar.pdf +2 -2
  37. data/processed/RQ1/RadarCharts/MarkdownValidator_radar.pdf +2 -2
  38. data/processed/RQ1/RadarCharts/Model_Legend.pdf +1 -1
  39. data/processed/RQ1/RadarCharts/Model_Legend_horizontal.pdf +1 -1
  40. data/processed/RQ1/RadarCharts/Overall_Average_radar.pdf +2 -2
  41. data/processed/RQ1/RadarCharts/RecruitmentAssistant-A2A_radar.pdf +2 -2
  42. data/processed/RQ1/RadarCharts/RecruitmentAssistant-H_A2A_radar.pdf +2 -2
  43. data/processed/RQ1/RadarCharts/RecruitmentAssistant-MCP_radar.pdf +2 -2
  44. data/processed/RQ1/RadarCharts/RecruitmentAssistant_Series_Average_radar.pdf +1 -1
  45. data/processed/RQ1/RadarCharts/SQLAssistant-A2A_radar.pdf +2 -2
  46. data/processed/RQ1/RadarCharts/SQLAssistant-H_A2A_radar.pdf +1 -1
  47. data/processed/RQ1/RadarCharts/SQLAssistant-MCP_radar.pdf +2 -2
  48. data/processed/RQ1/RadarCharts/SQLAssistant_Series_Average_radar.pdf +2 -2
  49. data/processed/RQ1/RadarCharts/SocialMediaManager-A2A_radar.pdf +2 -2
  50. data/processed/RQ1/RadarCharts/SocialMediaManager-H_A2A_radar.pdf +2 -2
data/processed/RQ1/EmailResponder-MCP/evaluate_success.py CHANGED
@@ -115,7 +115,7 @@ def save_detailed_results(results, output_file="success-finish_detailed_results.
115
  print(f"\n[OK] Detailed results saved to: {output_path}")
116
 
117
 
118
- def save_csv_summary(results, output_file="success-finish_rate.csv"):
119
  """Save summary to CSV file."""
120
  import csv
121
 
 
115
  print(f"\n[OK] Detailed results saved to: {output_path}")
116
 
117
 
118
+ def save_csv_summary(results, output_file="success_rate.csv"):
119
  """Save summary to CSV file."""
120
  import csv
121
 
data/processed/RQ1/EmailResponder-MCP/{success-finish_rate.csv → success_rate.csv} RENAMED
File without changes
data/processed/RQ1/EmailResponder/evaluate_success.py CHANGED
@@ -114,7 +114,7 @@ def save_detailed_results(results, output_file="success-finish_detailed_results.
114
  print(f"\n✅ Detailed results saved to: {output_path}")
115
 
116
 
117
- def save_csv_summary(results, output_file="success-finish_rate.csv"):
118
  """Save summary to CSV file"""
119
  import csv
120
 
 
114
  print(f"\n✅ Detailed results saved to: {output_path}")
115
 
116
 
117
+ def save_csv_summary(results, output_file="success_rate.csv"):
118
  """Save summary to CSV file"""
119
  import csv
120
 
data/processed/RQ1/EmailResponder/{success-finish_rate.csv → success_rate.csv} RENAMED
File without changes
data/processed/RQ1/GameBuilder-MCP/evaluate_success.py CHANGED
@@ -115,7 +115,7 @@ def save_detailed_results(results, output_file="success-finish_detailed_results.
115
  print(f"\n✅ Detailed results saved to: {output_path}")
116
 
117
 
118
- def save_csv_summary(results, output_file="success-finish_rate.csv"):
119
  """Save a summary to a CSV file."""
120
  import csv
121
 
 
115
  print(f"\n✅ Detailed results saved to: {output_path}")
116
 
117
 
118
+ def save_csv_summary(results, output_file="success_rate.csv"):
119
  """Save a summary to a CSV file."""
120
  import csv
121
 
data/processed/RQ1/GameBuilder-MCP/{success-finish_rate.csv → success_rate.csv} RENAMED
File without changes
data/processed/RQ1/GameBuilder/evaluate_success.py CHANGED
@@ -115,7 +115,7 @@ def save_detailed_results(results, output_file="success-finish_detailed_results.
115
  print(f"\n✅ Detailed results saved to: {output_path}")
116
 
117
 
118
- def save_csv_summary(results, output_file="success-finish_rate.csv"):
119
  """Save summary to CSV file"""
120
  import csv
121
 
 
115
  print(f"\n✅ Detailed results saved to: {output_path}")
116
 
117
 
118
+ def save_csv_summary(results, output_file="success_rate.csv"):
119
  """Save summary to CSV file"""
120
  import csv
121
 
data/processed/RQ1/GameBuilder/{success-finish_rate.csv → success_rate.csv} RENAMED
File without changes
data/processed/RQ1/MarkdownValidator-MCP/evaluate_success.py CHANGED
@@ -115,7 +115,7 @@ def save_detailed_results(results, output_file="success-finish_detailed_results.
115
  print(f"\nDetailed results saved to: {output_path}")
116
 
117
 
118
- def save_csv_summary(results, output_file="success-finish_rate.csv"):
119
  """Save summary to CSV file"""
120
  import csv
121
 
 
115
  print(f"\nDetailed results saved to: {output_path}")
116
 
117
 
118
+ def save_csv_summary(results, output_file="success_rate.csv"):
119
  """Save summary to CSV file"""
120
  import csv
121
 
data/processed/RQ1/MarkdownValidator-MCP/{success-finish_rate.csv → success_rate.csv} RENAMED
File without changes
data/processed/RQ1/MarkdownValidator/evaluate_success.py CHANGED
@@ -115,7 +115,7 @@ def save_detailed_results(results, output_file="success-finish_detailed_results.
115
  print(f"\nDetailed results saved: {output_path}")
116
 
117
 
118
- def save_csv_summary(results, output_file="success-finish_rate.csv"):
119
  """Save summary to CSV file"""
120
  import csv
121
 
 
115
  print(f"\nDetailed results saved: {output_path}")
116
 
117
 
118
+ def save_csv_summary(results, output_file="success_rate.csv"):
119
  """Save summary to CSV file"""
120
  import csv
121
 
data/processed/RQ1/MarkdownValidator/{success-finish_rate.csv → success_rate.csv} RENAMED
File without changes
data/processed/RQ1/RadarCharts/All_Open_vs_Closed_summaries.md CHANGED
@@ -2,62 +2,62 @@
2
 
3
  | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
4
  | --- | --- | --- | --- | --- | --- | --- | --- |
5
- | open | 3 | 0.21 | 0.44 | 0.84 | 0.93 | 0.18 | 0.65 |
6
- | closed | 4 | 0.13 | 0.24 | 0.91 | 0.81 | 0.12 | 0.56 |
7
 
8
  # BookWriter Open vs Closed Models
9
 
10
- | Series | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
11
- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
12
- | BookWriter | open | 3 | 0.01 | 0.20 | 0.91 | 0.86 | 0.04 | 1.00 |
13
- | BookWriter | closed | 4 | 0.00 | 0.00 | 0.94 | 0.66 | 0.00 | 1.00 |
14
 
15
  # EmailResponder Open vs Closed Models
16
 
17
- | Series | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
18
- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
19
- | EmailResponder | open | 3 | 0.07 | 0.11 | 0.92 | 0.87 | 0.06 | 0.00 |
20
- | EmailResponder | closed | 4 | 0.12 | 0.14 | 0.98 | 0.82 | 0.05 | 0.00 |
21
 
22
  # GameBuilder Open vs Closed Models
23
 
24
- | Series | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
25
- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
26
- | GameBuilder | open | 3 | 0.89 | 0.97 | 0.98 | 0.99 | 0.00 | 0.00 |
27
- | GameBuilder | closed | 4 | 0.47 | 0.52 | 0.99 | 0.89 | 0.01 | 0.00 |
28
 
29
  # LandingPageGenerator Open vs Closed Models
30
 
31
- | Series | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | average_score |
32
- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
33
- | LandingPageGenerator | open | 3 | 0.11 | 0.53 | 0.83 | 0.97 | 0.04 | 0.54 |
34
- | LandingPageGenerator | closed | 4 | 0.08 | 0.35 | 0.89 | 0.87 | 0.01 | 0.29 |
35
 
36
  # MarkdownValidator Open vs Closed Models
37
 
38
- | Series | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | average_score |
39
- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
40
- | MarkdownValidator | open | 3 | 0.99 | 0.99 | 1.00 | 1.00 | 0.00 | 0.82 |
41
- | MarkdownValidator | closed | 4 | 0.59 | 0.61 | 0.96 | 0.90 | 0.07 | 0.44 |
42
 
43
  # RecruitmentAssistant Open vs Closed Models
44
 
45
- | Series | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
46
- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
47
- | RecruitmentAssistant | open | 3 | 0.01 | 0.31 | 0.81 | 0.94 | 0.10 | 0.99 |
48
- | RecruitmentAssistant | closed | 4 | 0.00 | 0.17 | 0.92 | 0.83 | 0.00 | 1.00 |
49
 
50
  # SQLAssistant Open vs Closed Models
51
 
52
- | Series | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | average_score |
53
- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
54
- | SQLAssistant | open | 3 | 0.00 | 0.15 | 0.66 | 0.91 | 0.59 | 0.47 |
55
- | SQLAssistant | closed | 4 | 0.00 | 0.06 | 0.79 | 0.83 | 0.46 | 0.38 |
56
 
57
  # SocialMediaManager Open vs Closed Models
58
 
59
- | Series | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
60
- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
61
- | SocialMediaManager | open | 3 | 0.06 | 0.50 | 0.75 | 0.92 | 0.45 | 0.98 |
62
- | SocialMediaManager | closed | 4 | 0.03 | 0.27 | 0.87 | 0.72 | 0.29 | 0.99 |
63
 
 
2
 
3
  | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
4
  | --- | --- | --- | --- | --- | --- | --- | --- |
5
+ | open | 3 | 0.27 | 0.47 | 0.86 | 0.93 | 0.16 | 0.99 |
6
+ | closed | 4 | 0.16 | 0.27 | 0.92 | 0.82 | 0.11 | 1.00 |
7
 
8
  # BookWriter Open vs Closed Models
9
 
10
+ | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
11
+ | --- | --- | --- | --- | --- | --- | --- | --- |
12
+ | open | 3 | 0.01 | 0.20 | 0.91 | 0.86 | 0.04 | 1.00 |
13
+ | closed | 4 | 0.00 | 0.00 | 0.94 | 0.66 | 0.00 | 1.00 |
14
 
15
  # EmailResponder Open vs Closed Models
16
 
17
+ | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
18
+ | --- | --- | --- | --- | --- | --- | --- | --- |
19
+ | open | 3 | 0.07 | 0.11 | 0.92 | 0.87 | 0.06 | 1.00 |
20
+ | closed | 4 | 0.12 | 0.14 | 0.98 | 0.82 | 0.05 | 1.00 |
21
 
22
  # GameBuilder Open vs Closed Models
23
 
24
+ | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
25
+ | --- | --- | --- | --- | --- | --- | --- | --- |
26
+ | open | 3 | 0.89 | 0.97 | 0.98 | 0.99 | 0.00 | 0.99 |
27
+ | closed | 4 | 0.47 | 0.52 | 0.99 | 0.89 | 0.01 | 1.00 |
28
 
29
  # LandingPageGenerator Open vs Closed Models
30
 
31
+ | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | average_score |
32
+ | --- | --- | --- | --- | --- | --- | --- | --- |
33
+ | open | 3 | 0.11 | 0.53 | 0.83 | 0.97 | 0.04 | 0.54 |
34
+ | closed | 4 | 0.08 | 0.35 | 0.89 | 0.87 | 0.01 | 0.29 |
35
 
36
  # MarkdownValidator Open vs Closed Models
37
 
38
+ | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | average_score |
39
+ | --- | --- | --- | --- | --- | --- | --- | --- |
40
+ | open | 3 | 0.99 | 0.99 | 1.00 | 1.00 | 0.00 | 0.82 |
41
+ | closed | 4 | 0.59 | 0.61 | 0.96 | 0.90 | 0.07 | 0.44 |
42
 
43
  # RecruitmentAssistant Open vs Closed Models
44
 
45
+ | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
46
+ | --- | --- | --- | --- | --- | --- | --- | --- |
47
+ | open | 3 | 0.01 | 0.31 | 0.81 | 0.94 | 0.10 | 0.99 |
48
+ | closed | 4 | 0.00 | 0.17 | 0.92 | 0.83 | 0.00 | 1.00 |
49
 
50
  # SQLAssistant Open vs Closed Models
51
 
52
+ | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | average_score |
53
+ | --- | --- | --- | --- | --- | --- | --- | --- |
54
+ | open | 3 | 0.00 | 0.15 | 0.66 | 0.91 | 0.59 | 0.47 |
55
+ | closed | 4 | 0.00 | 0.06 | 0.79 | 0.83 | 0.46 | 0.38 |
56
 
57
  # SocialMediaManager Open vs Closed Models
58
 
59
+ | Group | Model Count | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
60
+ | --- | --- | --- | --- | --- | --- | --- | --- |
61
+ | open | 3 | 0.06 | 0.50 | 0.75 | 0.92 | 0.45 | 0.98 |
62
+ | closed | 4 | 0.03 | 0.27 | 0.87 | 0.72 | 0.29 | 0.99 |
63
 
data/processed/RQ1/RadarCharts/All_Series_Average_tables.md CHANGED
@@ -1,107 +1,125 @@
1
- # Overall Summary (All Projects)
2
 
3
- | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
4
- | --- | --- | --- | --- | --- | --- | --- |
5
- | GPT-5 | 0.24 | 0.31 | 0.93 | 0.90 | 0.11 | 0.63 |
6
- | GPT-4o-mini | 0.24 | 0.61 | 0.83 | 0.95 | 0.12 | 0.67 |
7
- | DeepSeek-V3-1 | 0.26 | 0.64 | 0.86 | 0.96 | 0.13 | 0.68 |
8
- | DeepSeek-R1 | 0.18 | 0.26 | 0.79 | 0.90 | 0.25 | 0.59 |
9
- | Gemini-2.5-flash | 0.03 | 0.03 | 0.93 | 0.69 | 0.14 | 0.48 |
10
- | Gemini-2.5-flash-nothinking | 0.02 | 0.02 | 0.94 | 0.70 | 0.12 | 0.48 |
11
- | Qwen3-235b | 0.20 | 0.41 | 0.88 | 0.92 | 0.16 | 0.67 |
 
 
12
 
13
  # BookWriter Series Average
14
 
 
 
15
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
16
  | --- | --- | --- | --- | --- | --- | --- |
17
- | GPT-5 | 0.00 | 0.00 | 0.97 | 0.77 | 0.00 | 1.00 |
18
- | GPT-4o-mini | 0.00 | 0.00 | 0.85 | 0.83 | 0.00 | 1.00 |
19
- | DeepSeek-V3-1 | 0.04 | 0.58 | 0.89 | 0.98 | 0.01 | 1.00 |
20
  | DeepSeek-R1 | 0.00 | 0.01 | 0.88 | 0.81 | 0.09 | 1.00 |
 
 
 
21
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.98 | 0.52 | 0.00 | 1.00 |
22
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.98 | 0.53 | 0.01 | 1.00 |
23
  | Qwen3-235b | 0.00 | 0.00 | 0.95 | 0.79 | 0.01 | 1.00 |
24
 
25
  # EmailResponder Series Average
26
 
 
 
27
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
28
  | --- | --- | --- | --- | --- | --- | --- |
29
- | GPT-5 | 0.01 | 0.01 | 1.00 | 0.83 | 0.00 | 0.00 |
30
- | GPT-4o-mini | 0.43 | 0.47 | 0.99 | 0.91 | 0.00 | 0.00 |
31
- | DeepSeek-V3-1 | 0.16 | 0.24 | 0.98 | 0.88 | 0.02 | 0.00 |
32
- | DeepSeek-R1 | 0.00 | 0.03 | 0.77 | 0.89 | 0.16 | 0.00 |
33
- | Gemini-2.5-flash | 0.01 | 0.01 | 0.97 | 0.74 | 0.10 | 0.00 |
34
- | Gemini-2.5-flash-nothinking | 0.05 | 0.06 | 0.96 | 0.81 | 0.12 | 0.00 |
35
- | Qwen3-235b | 0.06 | 0.07 | 1.00 | 0.85 | 0.00 | 0.00 |
36
 
37
  # GameBuilder Series Average
38
 
 
 
39
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
40
  | --- | --- | --- | --- | --- | --- | --- |
41
- | GPT-5 | 1.00 | 1.00 | 1.00 | 1.00 | 0.00 | 0.00 |
42
- | GPT-4o-mini | 0.88 | 1.00 | 0.96 | 1.00 | 0.00 | 0.00 |
43
- | DeepSeek-V3-1 | 0.90 | 1.00 | 0.97 | 1.00 | 0.00 | 0.00 |
44
- | DeepSeek-R1 | 0.94 | 0.96 | 1.00 | 0.99 | 0.00 | 0.00 |
45
- | Gemini-2.5-flash | 0.00 | 0.00 | 1.00 | 0.75 | 0.00 | 0.00 |
46
- | Gemini-2.5-flash-nothinking | 0.02 | 0.06 | 0.99 | 0.80 | 0.04 | 0.00 |
47
- | Qwen3-235b | 0.84 | 0.94 | 0.99 | 0.99 | 0.00 | 0.00 |
48
 
49
  # LandingPageGenerator Series Average
50
 
 
 
51
  | model | exact_match | any_order_match | precision | recall | retry_rate | average_score |
52
  | --- | --- | --- | --- | --- | --- | --- |
53
- | GPT-5 | 0.29 | 0.42 | 0.86 | 0.97 | 0.00 | 0.47 |
54
- | GPT-4o-mini | 0.04 | 0.98 | 0.78 | 1.00 | 0.00 | 0.67 |
55
- | DeepSeek-V3-1 | 0.24 | 0.99 | 0.86 | 1.00 | 0.01 | 0.86 |
56
  | DeepSeek-R1 | 0.00 | 0.13 | 0.78 | 0.94 | 0.06 | 0.25 |
 
 
 
57
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.97 | 0.76 | 0.00 | 0.00 |
58
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.95 | 0.77 | 0.03 | 0.01 |
59
  | Qwen3-235b | 0.09 | 0.45 | 0.84 | 0.97 | 0.03 | 0.52 |
60
 
61
  # MarkdownValidator Series Average
62
 
 
 
63
  | model | exact_match | any_order_match | precision | recall | retry_rate | average_score |
64
  | --- | --- | --- | --- | --- | --- | --- |
65
- | GPT-5 | 1.00 | 1.00 | 1.00 | 1.00 | 0.00 | 0.71 |
66
- | GPT-4o-mini | 0.97 | 1.00 | 0.99 | 1.00 | 0.00 | 0.82 |
67
- | DeepSeek-V3-1 | 1.00 | 1.00 | 1.00 | 1.00 | 0.00 | 0.84 |
68
  | DeepSeek-R1 | 0.98 | 1.00 | 1.00 | 1.00 | 0.00 | 0.81 |
 
 
 
69
  | Gemini-2.5-flash | 0.26 | 0.33 | 0.87 | 0.88 | 0.21 | 0.08 |
70
  | Gemini-2.5-flash-nothinking | 0.13 | 0.13 | 0.96 | 0.73 | 0.06 | 0.14 |
71
  | Qwen3-235b | 0.98 | 0.98 | 1.00 | 0.99 | 0.00 | 0.81 |
72
 
73
  # RecruitmentAssistant Series Average
74
 
 
 
75
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
76
  | --- | --- | --- | --- | --- | --- | --- |
77
- | GPT-5 | 0.00 | 0.00 | 0.98 | 0.90 | 0.00 | 1.00 |
78
- | GPT-4o-mini | 0.00 | 0.70 | 0.76 | 0.98 | 0.00 | 1.00 |
79
- | DeepSeek-V3-1 | 0.00 | 0.36 | 0.84 | 0.97 | 0.00 | 1.00 |
80
  | DeepSeek-R1 | 0.00 | 0.11 | 0.84 | 0.90 | 0.05 | 0.99 |
 
 
 
81
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.98 | 0.69 | 0.00 | 1.00 |
82
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.97 | 0.74 | 0.00 | 1.00 |
83
  | Qwen3-235b | 0.02 | 0.46 | 0.75 | 0.96 | 0.26 | 0.98 |
84
 
85
  # SQLAssistant Series Average
86
 
 
 
87
  | model | exact_match | any_order_match | precision | recall | retry_rate | average_score |
88
  | --- | --- | --- | --- | --- | --- | --- |
89
- | GPT-5 | 0.00 | 0.01 | 0.82 | 0.86 | 0.44 | 0.45 |
90
- | GPT-4o-mini | 0.00 | 0.22 | 0.77 | 0.94 | 0.37 | 0.52 |
91
- | DeepSeek-V3-1 | 0.00 | 0.17 | 0.66 | 0.90 | 0.61 | 0.37 |
92
  | DeepSeek-R1 | 0.00 | 0.04 | 0.52 | 0.88 | 0.80 | 0.40 |
 
 
 
93
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.76 | 0.77 | 0.57 | 0.29 |
94
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.82 | 0.76 | 0.48 | 0.27 |
95
  | Qwen3-235b | 0.00 | 0.24 | 0.80 | 0.95 | 0.37 | 0.64 |
96
 
97
  # SocialMediaManager Series Average
98
 
 
 
99
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
100
  | --- | --- | --- | --- | --- | --- | --- |
101
- | GPT-5 | 0.02 | 0.37 | 0.86 | 0.88 | 0.34 | 0.98 |
102
- | GPT-4o-mini | 0.12 | 0.71 | 0.72 | 0.97 | 0.47 | 0.96 |
103
- | DeepSeek-V3-1 | 0.17 | 0.87 | 0.78 | 0.97 | 0.27 | 1.00 |
104
  | DeepSeek-R1 | 0.01 | 0.22 | 0.64 | 0.88 | 0.62 | 0.94 |
 
 
 
105
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.94 | 0.52 | 0.21 | 1.00 |
106
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.96 | 0.51 | 0.15 | 1.00 |
107
  | Qwen3-235b | 0.02 | 0.42 | 0.82 | 0.89 | 0.46 | 0.98 |
 
1
+ # Overall Summary (All Projects - Weighted Average)
2
 
3
+ _Note: average_score is only computed for GT projects (MarkdownValidator, SQLAssistant, LandingPageGenerator). pass_rate is only computed for non-GT projects._
4
+
5
+ | model | exact_match | any_order_match | precision | recall | retry_rate | average_score | pass_rate |
6
+ | --- | --- | --- | --- | --- | --- | --- | --- |
7
+ | DeepSeek-R1 | 0.13 | 0.21 | 0.77 | 0.90 | 0.26 | 0.45 | 0.98 |
8
+ | DeepSeek-V3-1 | 0.22 | 0.59 | 0.86 | 0.96 | 0.13 | 0.67 | 1.00 |
9
+ | GPT-4o-mini | 0.22 | 0.58 | 0.84 | 0.95 | 0.12 | 0.65 | 0.99 |
10
+ | GPT-5 | 0.18 | 0.25 | 0.93 | 0.89 | 0.11 | 0.52 | 1.00 |
11
+ | Gemini-2.5-flash | 0.03 | 0.03 | 0.93 | 0.69 | 0.14 | 0.13 | 1.00 |
12
+ | Gemini-2.5-flash-nothinking | 0.02 | 0.02 | 0.94 | 0.70 | 0.12 | 0.14 | 1.00 |
13
+ | Qwen3-235b | 0.16 | 0.36 | 0.88 | 0.91 | 0.16 | 0.64 | 0.99 |
14
 
15
  # BookWriter Series Average
16
 
17
+ _This is a non-GT project series. Using pass_rate from success_rate.csv._
18
+
19
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
20
  | --- | --- | --- | --- | --- | --- | --- |
 
 
 
21
  | DeepSeek-R1 | 0.00 | 0.01 | 0.88 | 0.81 | 0.09 | 1.00 |
22
+ | DeepSeek-V3-1 | 0.04 | 0.58 | 0.89 | 0.98 | 0.01 | 1.00 |
23
+ | GPT-4o-mini | 0.00 | 0.00 | 0.85 | 0.83 | 0.00 | 1.00 |
24
+ | GPT-5 | 0.00 | 0.00 | 0.97 | 0.77 | 0.00 | 1.00 |
25
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.98 | 0.52 | 0.00 | 1.00 |
26
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.98 | 0.53 | 0.01 | 1.00 |
27
  | Qwen3-235b | 0.00 | 0.00 | 0.95 | 0.79 | 0.01 | 1.00 |
28
 
29
  # EmailResponder Series Average
30
 
31
+ _This is a non-GT project series. Using pass_rate from success_rate.csv._
32
+
33
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
34
  | --- | --- | --- | --- | --- | --- | --- |
35
+ | DeepSeek-R1 | 0.00 | 0.03 | 0.77 | 0.89 | 0.16 | 0.99 |
36
+ | DeepSeek-V3-1 | 0.16 | 0.24 | 0.98 | 0.88 | 0.02 | 1.00 |
37
+ | GPT-4o-mini | 0.43 | 0.47 | 0.99 | 0.91 | 0.00 | 1.00 |
38
+ | GPT-5 | 0.01 | 0.01 | 1.00 | 0.83 | 0.00 | 1.00 |
39
+ | Gemini-2.5-flash | 0.01 | 0.01 | 0.97 | 0.74 | 0.10 | 0.99 |
40
+ | Gemini-2.5-flash-nothinking | 0.04 | 0.06 | 0.96 | 0.81 | 0.12 | 1.00 |
41
+ | Qwen3-235b | 0.06 | 0.07 | 1.00 | 0.85 | 0.00 | 1.00 |
42
 
43
  # GameBuilder Series Average
44
 
45
+ _This is a non-GT project series. Using pass_rate from success_rate.csv._
46
+
47
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
48
  | --- | --- | --- | --- | --- | --- | --- |
49
+ | DeepSeek-R1 | 0.94 | 0.96 | 1.00 | 0.99 | 0.00 | 1.00 |
50
+ | DeepSeek-V3-1 | 0.90 | 1.00 | 0.97 | 1.00 | 0.00 | 0.98 |
51
+ | GPT-4o-mini | 0.88 | 1.00 | 0.96 | 1.00 | 0.00 | 1.00 |
52
+ | GPT-5 | 1.00 | 1.00 | 1.00 | 1.00 | 0.00 | 1.00 |
53
+ | Gemini-2.5-flash | 0.00 | 0.00 | 1.00 | 0.75 | 0.00 | 1.00 |
54
+ | Gemini-2.5-flash-nothinking | 0.02 | 0.06 | 0.99 | 0.80 | 0.04 | 1.00 |
55
+ | Qwen3-235b | 0.84 | 0.94 | 0.99 | 0.99 | 0.00 | 1.00 |
56
 
57
  # LandingPageGenerator Series Average
58
 
59
+ _This is a GT project series. Using average_score from score_summary.csv._
60
+
61
  | model | exact_match | any_order_match | precision | recall | retry_rate | average_score |
62
  | --- | --- | --- | --- | --- | --- | --- |
 
 
 
63
  | DeepSeek-R1 | 0.00 | 0.13 | 0.78 | 0.94 | 0.06 | 0.25 |
64
+ | DeepSeek-V3-1 | 0.24 | 0.99 | 0.86 | 1.00 | 0.01 | 0.86 |
65
+ | GPT-4o-mini | 0.04 | 0.98 | 0.78 | 1.00 | 0.00 | 0.67 |
66
+ | GPT-5 | 0.29 | 0.42 | 0.86 | 0.97 | 0.00 | 0.47 |
67
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.97 | 0.76 | 0.00 | 0.00 |
68
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.95 | 0.77 | 0.03 | 0.01 |
69
  | Qwen3-235b | 0.09 | 0.45 | 0.84 | 0.97 | 0.03 | 0.52 |
70
 
71
  # MarkdownValidator Series Average
72
 
73
+ _This is a GT project series. Using average_score from score_summary.csv._
74
+
75
  | model | exact_match | any_order_match | precision | recall | retry_rate | average_score |
76
  | --- | --- | --- | --- | --- | --- | --- |
 
 
 
77
  | DeepSeek-R1 | 0.98 | 1.00 | 1.00 | 1.00 | 0.00 | 0.81 |
78
+ | DeepSeek-V3-1 | 1.00 | 1.00 | 1.00 | 1.00 | 0.00 | 0.84 |
79
+ | GPT-4o-mini | 0.97 | 1.00 | 0.99 | 1.00 | 0.00 | 0.82 |
80
+ | GPT-5 | 1.00 | 1.00 | 1.00 | 1.00 | 0.00 | 0.71 |
81
  | Gemini-2.5-flash | 0.26 | 0.33 | 0.87 | 0.88 | 0.21 | 0.08 |
82
  | Gemini-2.5-flash-nothinking | 0.13 | 0.13 | 0.96 | 0.73 | 0.06 | 0.14 |
83
  | Qwen3-235b | 0.98 | 0.98 | 1.00 | 0.99 | 0.00 | 0.81 |
84
 
85
  # RecruitmentAssistant Series Average
86
 
87
+ _This is a non-GT project series. Using pass_rate from success_rate.csv._
88
+
89
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
90
  | --- | --- | --- | --- | --- | --- | --- |
 
 
 
91
  | DeepSeek-R1 | 0.00 | 0.11 | 0.84 | 0.90 | 0.05 | 0.99 |
92
+ | DeepSeek-V3-1 | 0.00 | 0.36 | 0.84 | 0.97 | 0.00 | 1.00 |
93
+ | GPT-4o-mini | 0.00 | 0.70 | 0.76 | 0.98 | 0.00 | 1.00 |
94
+ | GPT-5 | 0.00 | 0.00 | 0.98 | 0.90 | 0.00 | 1.00 |
95
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.98 | 0.69 | 0.00 | 1.00 |
96
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.97 | 0.74 | 0.00 | 1.00 |
97
  | Qwen3-235b | 0.02 | 0.46 | 0.75 | 0.96 | 0.26 | 0.98 |
98
 
99
  # SQLAssistant Series Average
100
 
101
+ _This is a GT project series. Using average_score from score_summary.csv._
102
+
103
  | model | exact_match | any_order_match | precision | recall | retry_rate | average_score |
104
  | --- | --- | --- | --- | --- | --- | --- |
 
 
 
105
  | DeepSeek-R1 | 0.00 | 0.04 | 0.52 | 0.88 | 0.80 | 0.40 |
106
+ | DeepSeek-V3-1 | 0.00 | 0.17 | 0.66 | 0.90 | 0.61 | 0.37 |
107
+ | GPT-4o-mini | 0.00 | 0.22 | 0.77 | 0.94 | 0.37 | 0.52 |
108
+ | GPT-5 | 0.00 | 0.01 | 0.82 | 0.86 | 0.44 | 0.45 |
109
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.76 | 0.77 | 0.57 | 0.29 |
110
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.82 | 0.76 | 0.48 | 0.27 |
111
  | Qwen3-235b | 0.00 | 0.24 | 0.80 | 0.95 | 0.37 | 0.64 |
112
 
113
  # SocialMediaManager Series Average
114
 
115
+ _This is a non-GT project series. Using pass_rate from success_rate.csv._
116
+
117
  | model | exact_match | any_order_match | precision | recall | retry_rate | pass_rate |
118
  | --- | --- | --- | --- | --- | --- | --- |
 
 
 
119
  | DeepSeek-R1 | 0.01 | 0.22 | 0.64 | 0.88 | 0.62 | 0.94 |
120
+ | DeepSeek-V3-1 | 0.17 | 0.87 | 0.78 | 0.97 | 0.27 | 1.00 |
121
+ | GPT-4o-mini | 0.12 | 0.71 | 0.72 | 0.97 | 0.47 | 0.96 |
122
+ | GPT-5 | 0.02 | 0.37 | 0.86 | 0.88 | 0.34 | 0.98 |
123
  | Gemini-2.5-flash | 0.00 | 0.00 | 0.94 | 0.52 | 0.21 | 1.00 |
124
  | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | 0.96 | 0.51 | 0.15 | 1.00 |
125
  | Qwen3-235b | 0.02 | 0.42 | 0.82 | 0.89 | 0.46 | 0.98 |
data/processed/RQ1/RadarCharts/BookWriter-A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9eaed441462082800ee22bfb8ef5a1cc84ab4c7f9a1a956570ad6fe986e7034d
3
- size 18128
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4da0e9d550b5ba06e37a72a279ac4fbdd6a2a4cda6ff704d37aa17907569d41c
3
+ size 18417
data/processed/RQ1/RadarCharts/BookWriter-H_A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:747927181126917ca0e6e331f60b03c32440bcaf92d46c0f5bc6e884b8b81d37
3
- size 18037
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27a597245cd6396c7f09be84845e95571d6207e13c19c51dfd12287a02d6eab8
3
+ size 18321
data/processed/RQ1/RadarCharts/BookWriter-MCP_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1df13bbe21a802b24c81eaeb1884f69994df008f757f94abf545611e0cb94001
3
- size 18171
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c212b19b2b232560f4be82708dcc3627d4fd1672f1a6f17bf706fc4b84685fae
3
+ size 18476
data/processed/RQ1/RadarCharts/BookWriter_Series_Average_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8b72a4c02471ff93ea36ece48d9907dafba89e32d9d5e15387c1d4ca000d45e9
3
  size 18189
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79c4c09f9bf7c088d0604cc23cad260c52eb608b0033805f0664ab62d0e8d3dc
3
  size 18189
data/processed/RQ1/RadarCharts/Comparison_A2A_vs_A2A_mix.md DELETED
@@ -1,68 +0,0 @@
1
- # Overall A2A vs A2A_mix Comparison
2
-
3
- Averaged across all projects: SQL_assistant, intelligent_recruitment_platform, landing_page_generator, self_evaluation_loop_flow, write_a_book_with_flows
4
-
5
- ## Framework-Level Comparison (All Models Averaged)
6
-
7
- | Metric | A2A | A2A_mix | Diff (A2A-A2A_mix) |
8
- | --- | --- | --- | --- |
9
-
10
- ---
11
-
12
- ## exact_match
13
-
14
- | Model | A2A (Avg) | A2A_mix (Avg) | Diff (A2A-A2A_mix) |
15
- | --- | --- | --- | --- |
16
-
17
- ## any_order_match
18
-
19
- | Model | A2A (Avg) | A2A_mix (Avg) | Diff (A2A-A2A_mix) |
20
- | --- | --- | --- | --- |
21
-
22
- ## precision
23
-
24
- | Model | A2A (Avg) | A2A_mix (Avg) | Diff (A2A-A2A_mix) |
25
- | --- | --- | --- | --- |
26
-
27
- ## recall
28
-
29
- | Model | A2A (Avg) | A2A_mix (Avg) | Diff (A2A-A2A_mix) |
30
- | --- | --- | --- | --- |
31
-
32
- ## retry_rate
33
-
34
- | Model | A2A (Avg) | A2A_mix (Avg) | Diff (A2A-A2A_mix) |
35
- | --- | --- | --- | --- |
36
-
37
- ## average_score
38
-
39
- | Model | A2A (Avg) | A2A_mix (Avg) | Diff (A2A-A2A_mix) |
40
- | --- | --- | --- | --- |
41
-
42
- ## pass_rate
43
-
44
- | Model | A2A (Avg) | A2A_mix (Avg) | Diff (A2A-A2A_mix) |
45
- | --- | --- | --- | --- |
46
-
47
- ---
48
-
49
- # SQL_assistant: A2A vs A2A_mix
50
-
51
- _Data not available for comparison_
52
-
53
- # intelligent_recruitment_platform: A2A vs A2A_mix
54
-
55
- _Data not available for comparison_
56
-
57
- # landing_page_generator: A2A vs A2A_mix
58
-
59
- _Data not available for comparison_
60
-
61
- # self_evaluation_loop_flow: A2A vs A2A_mix
62
-
63
- _Data not available for comparison_
64
-
65
- # write_a_book_with_flows: A2A vs A2A_mix
66
-
67
- _Data not available for comparison_
68
-
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
data/processed/RQ1/RadarCharts/Comparison_A2A_vs_H_A2A.md ADDED
@@ -0,0 +1,539 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Overall A2A vs H_A2A Comparison
2
+
3
+ Averaged across all projects: BookWriter, LandingPageGenerator, RecruitmentAssistant, SQLAssistant, SocialMediaManager
4
+
5
+ ## Framework-Level Comparison (All Models Averaged)
6
+
7
+ | Metric | A2A | H_A2A | Diff (A2A-H_A2A) |
8
+ | --- | --- | --- | --- |
9
+ | exact_match | 0.01 | 0.01 | -0.01 |
10
+ | any_order_match | 0.25 | 0.19 | +0.05 |
11
+ | precision | 0.82 | 0.85 | -0.03 |
12
+ | recall | 0.84 | 0.87 | -0.03 |
13
+ | retry_rate | 0.22 | 0.14 | +0.08 |
14
+ | average_score | 0.31 | 0.45 | -0.14 |
15
+ | pass_rate | 0.99 | 1.00 | -0.01 |
16
+
17
+ ---
18
+
19
+ ## exact_match
20
+
21
+ | Model | A2A (Avg) | H_A2A (Avg) | Diff (A2A-H_A2A) |
22
+ | --- | --- | --- | --- |
23
+ | DeepSeek-R1 | 0.00 | 0.00 | -0.00 |
24
+ | DeepSeek-V3-1 | 0.03 | 0.04 | -0.01 |
25
+ | GPT-4o-mini | 0.01 | 0.05 | -0.05 |
26
+ | GPT-5 | 0.01 | 0.00 | +0.01 |
27
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
28
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
29
+ | Qwen3-235b | 0.01 | 0.01 | +0.00 |
30
+
31
+ ## any_order_match
32
+
33
+ | Model | A2A (Avg) | H_A2A (Avg) | Diff (A2A-H_A2A) |
34
+ | --- | --- | --- | --- |
35
+ | DeepSeek-R1 | 0.05 | 0.12 | -0.07 |
36
+ | DeepSeek-V3-1 | 0.48 | 0.68 | -0.20 |
37
+ | GPT-4o-mini | 0.60 | 0.43 | +0.16 |
38
+ | GPT-5 | 0.17 | 0.03 | +0.14 |
39
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
40
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
41
+ | Qwen3-235b | 0.43 | 0.10 | +0.33 |
42
+
43
+ ## precision
44
+
45
+ | Model | A2A (Avg) | H_A2A (Avg) | Diff (A2A-H_A2A) |
46
+ | --- | --- | --- | --- |
47
+ | DeepSeek-R1 | 0.70 | 0.79 | -0.09 |
48
+ | DeepSeek-V3-1 | 0.74 | 0.84 | -0.11 |
49
+ | GPT-4o-mini | 0.79 | 0.81 | -0.03 |
50
+ | GPT-5 | 0.89 | 0.91 | -0.02 |
51
+ | Gemini-2.5-flash | 0.91 | 0.89 | +0.02 |
52
+ | Gemini-2.5-flash-nothinking | 0.93 | 0.90 | +0.03 |
53
+ | Qwen3-235b | 0.81 | 0.83 | -0.02 |
54
+
55
+ ## recall
56
+
57
+ | Model | A2A (Avg) | H_A2A (Avg) | Diff (A2A-H_A2A) |
58
+ | --- | --- | --- | --- |
59
+ | DeepSeek-R1 | 0.88 | 0.89 | -0.00 |
60
+ | DeepSeek-V3-1 | 0.95 | 0.96 | -0.01 |
61
+ | GPT-4o-mini | 0.96 | 0.92 | +0.04 |
62
+ | GPT-5 | 0.90 | 0.85 | +0.04 |
63
+ | Gemini-2.5-flash | 0.62 | 0.78 | -0.16 |
64
+ | Gemini-2.5-flash-nothinking | 0.64 | 0.78 | -0.14 |
65
+ | Qwen3-235b | 0.93 | 0.91 | +0.02 |
66
+
67
+ ## retry_rate
68
+
69
+ | Model | A2A (Avg) | H_A2A (Avg) | Diff (A2A-H_A2A) |
70
+ | --- | --- | --- | --- |
71
+ | DeepSeek-R1 | 0.37 | 0.26 | +0.11 |
72
+ | DeepSeek-V3-1 | 0.25 | 0.10 | +0.15 |
73
+ | GPT-4o-mini | 0.15 | 0.09 | +0.07 |
74
+ | GPT-5 | 0.18 | 0.06 | +0.11 |
75
+ | Gemini-2.5-flash | 0.19 | 0.18 | +0.01 |
76
+ | Gemini-2.5-flash-nothinking | 0.18 | 0.12 | +0.06 |
77
+ | Qwen3-235b | 0.25 | 0.20 | +0.05 |
78
+
79
+ ## average_score
80
+
81
+ | Model | A2A (Avg) | H_A2A (Avg) | Diff (A2A-H_A2A) |
82
+ | --- | --- | --- | --- |
83
+ | DeepSeek-R1 | 0.16 | 0.31 | -0.15 |
84
+ | DeepSeek-V3-1 | 0.49 | 0.70 | -0.21 |
85
+ | GPT-4o-mini | 0.57 | 0.58 | -0.01 |
86
+ | GPT-5 | 0.33 | 0.42 | -0.09 |
87
+ | Gemini-2.5-flash | 0.08 | 0.27 | -0.20 |
88
+ | Gemini-2.5-flash-nothinking | 0.03 | 0.29 | -0.26 |
89
+ | Qwen3-235b | 0.49 | 0.55 | -0.06 |
90
+
91
+ ## pass_rate
92
+
93
+ | Model | A2A (Avg) | H_A2A (Avg) | Diff (A2A-H_A2A) |
94
+ | --- | --- | --- | --- |
95
+ | DeepSeek-R1 | 0.97 | 0.99 | -0.02 |
96
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
97
+ | GPT-4o-mini | 0.98 | 1.00 | -0.02 |
98
+ | GPT-5 | 0.99 | 1.00 | -0.01 |
99
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
100
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
101
+ | Qwen3-235b | 0.99 | 0.99 | -0.01 |
102
+
103
+ ---
104
+
105
+ # BookWriter: A2A vs H_A2A
106
+
107
+ ## Overall Summary (Averaged Across All Models)
108
+
109
+ | Metric | A2A | H_A2A | Diff (A2A-H_A2A) |
110
+ | --- | --- | --- | --- |
111
+ | exact_match | 0.00 | 0.00 | +0.00 |
112
+ | any_order_match | 0.08 | 0.08 | -0.00 |
113
+ | precision | 0.93 | 0.91 | +0.02 |
114
+ | recall | 0.77 | 0.73 | +0.03 |
115
+ | retry_rate | 0.02 | 0.00 | +0.02 |
116
+ | pass_rate | 1.00 | 1.00 | +0.00 |
117
+
118
+ ---
119
+
120
+ ## exact_match
121
+
122
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
123
+ | --- | --- | --- | --- |
124
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
125
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
126
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
127
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
128
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
129
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
130
+ | Qwen3-235b | 0.00 | 0.00 | +0.00 |
131
+
132
+ ## any_order_match
133
+
134
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
135
+ | --- | --- | --- | --- |
136
+ | DeepSeek-R1 | 0.02 | 0.00 | +0.02 |
137
+ | DeepSeek-V3-1 | 0.53 | 0.58 | -0.05 |
138
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
139
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
140
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
141
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
142
+ | Qwen3-235b | 0.00 | 0.00 | +0.00 |
143
+
144
+ ## precision
145
+
146
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
147
+ | --- | --- | --- | --- |
148
+ | DeepSeek-R1 | 0.87 | 0.91 | -0.03 |
149
+ | DeepSeek-V3-1 | 0.79 | 0.92 | -0.13 |
150
+ | GPT-4o-mini | 0.90 | 0.80 | +0.10 |
151
+ | GPT-5 | 0.98 | 0.95 | +0.03 |
152
+ | Gemini-2.5-flash | 1.00 | 0.93 | +0.07 |
153
+ | Gemini-2.5-flash-nothinking | 1.00 | 0.93 | +0.07 |
154
+ | Qwen3-235b | 0.96 | 0.93 | +0.03 |
155
+
156
+ ## recall
157
+
158
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
159
+ | --- | --- | --- | --- |
160
+ | DeepSeek-R1 | 0.84 | 0.77 | +0.07 |
161
+ | DeepSeek-V3-1 | 0.99 | 0.95 | +0.04 |
162
+ | GPT-4o-mini | 0.86 | 0.77 | +0.09 |
163
+ | GPT-5 | 0.82 | 0.67 | +0.15 |
164
+ | Gemini-2.5-flash | 0.53 | 0.57 | -0.05 |
165
+ | Gemini-2.5-flash-nothinking | 0.53 | 0.57 | -0.04 |
166
+ | Qwen3-235b | 0.80 | 0.83 | -0.04 |
167
+
168
+ ## retry_rate
169
+
170
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
171
+ | --- | --- | --- | --- |
172
+ | DeepSeek-R1 | 0.13 | 0.02 | +0.12 |
173
+ | DeepSeek-V3-1 | 0.02 | 0.00 | +0.02 |
174
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
175
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
176
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
177
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
178
+ | Qwen3-235b | 0.00 | 0.00 | +0.00 |
179
+
180
+ ## pass_rate
181
+
182
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
183
+ | --- | --- | --- | --- |
184
+ | DeepSeek-R1 | 1.00 | 1.00 | +0.00 |
185
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
186
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
187
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
188
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
189
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
190
+ | Qwen3-235b | 1.00 | 1.00 | +0.00 |
191
+
192
+ # LandingPageGenerator: A2A vs H_A2A
193
+
194
+ ## Overall Summary (Averaged Across All Models)
195
+
196
+ | Metric | A2A | H_A2A | Diff (A2A-H_A2A) |
197
+ | --- | --- | --- | --- |
198
+ | exact_match | 0.00 | 0.00 | +0.00 |
199
+ | any_order_match | 0.40 | 0.29 | +0.11 |
200
+ | precision | 0.84 | 0.82 | +0.03 |
201
+ | recall | 0.91 | 0.93 | -0.02 |
202
+ | retry_rate | 0.02 | 0.01 | +0.01 |
203
+ | average_score | 0.30 | 0.33 | -0.03 |
204
+
205
+ ---
206
+
207
+ ## exact_match
208
+
209
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
210
+ | --- | --- | --- | --- |
211
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
212
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
213
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
214
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
215
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
216
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
217
+ | Qwen3-235b | 0.00 | 0.00 | +0.00 |
218
+
219
+ ## any_order_match
220
+
221
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
222
+ | --- | --- | --- | --- |
223
+ | DeepSeek-R1 | 0.05 | 0.03 | +0.02 |
224
+ | DeepSeek-V3-1 | 1.00 | 0.98 | +0.02 |
225
+ | GPT-4o-mini | 0.97 | 1.00 | -0.03 |
226
+ | GPT-5 | 0.27 | 0.00 | +0.27 |
227
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
228
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
229
+ | Qwen3-235b | 0.48 | 0.00 | +0.48 |
230
+
231
+ ## precision
232
+
233
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
234
+ | --- | --- | --- | --- |
235
+ | DeepSeek-R1 | 0.76 | 0.78 | -0.03 |
236
+ | DeepSeek-V3-1 | 0.83 | 0.77 | +0.06 |
237
+ | GPT-4o-mini | 0.74 | 0.78 | -0.03 |
238
+ | GPT-5 | 0.81 | 0.79 | +0.02 |
239
+ | Gemini-2.5-flash | 1.00 | 0.91 | +0.09 |
240
+ | Gemini-2.5-flash-nothinking | 0.98 | 0.87 | +0.12 |
241
+ | Qwen3-235b | 0.79 | 0.81 | -0.02 |
242
+
243
+ ## recall
244
+
245
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
246
+ | --- | --- | --- | --- |
247
+ | DeepSeek-R1 | 0.94 | 0.93 | +0.01 |
248
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
249
+ | GPT-4o-mini | 1.00 | 1.00 | -0.00 |
250
+ | GPT-5 | 0.97 | 0.94 | +0.03 |
251
+ | Gemini-2.5-flash | 0.74 | 0.87 | -0.13 |
252
+ | Gemini-2.5-flash-nothinking | 0.76 | 0.85 | -0.08 |
253
+ | Qwen3-235b | 0.98 | 0.94 | +0.04 |
254
+
255
+ ## retry_rate
256
+
257
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
258
+ | --- | --- | --- | --- |
259
+ | DeepSeek-R1 | 0.03 | 0.00 | +0.03 |
260
+ | DeepSeek-V3-1 | 0.00 | 0.03 | -0.03 |
261
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
262
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
263
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
264
+ | Gemini-2.5-flash-nothinking | 0.05 | 0.00 | +0.05 |
265
+ | Qwen3-235b | 0.05 | 0.05 | +0.00 |
266
+
267
+ ## average_score
268
+
269
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
270
+ | --- | --- | --- | --- |
271
+ | DeepSeek-R1 | 0.08 | 0.10 | -0.02 |
272
+ | DeepSeek-V3-1 | 0.86 | 0.88 | -0.02 |
273
+ | GPT-4o-mini | 0.57 | 0.59 | -0.02 |
274
+ | GPT-5 | 0.19 | 0.32 | -0.12 |
275
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
276
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
277
+ | Qwen3-235b | 0.38 | 0.41 | -0.03 |
278
+
279
+ # RecruitmentAssistant: A2A vs H_A2A
280
+
281
+ ## Overall Summary (Averaged Across All Models)
282
+
283
+ | Metric | A2A | H_A2A | Diff (A2A-H_A2A) |
284
+ | --- | --- | --- | --- |
285
+ | exact_match | 0.00 | 0.00 | +0.00 |
286
+ | any_order_match | 0.18 | 0.31 | -0.13 |
287
+ | precision | 0.89 | 0.87 | +0.02 |
288
+ | recall | 0.86 | 0.95 | -0.09 |
289
+ | retry_rate | 0.05 | 0.03 | +0.01 |
290
+ | pass_rate | 1.00 | 1.00 | -0.00 |
291
+
292
+ ---
293
+
294
+ ## exact_match
295
+
296
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
297
+ | --- | --- | --- | --- |
298
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
299
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
300
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
301
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
302
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
303
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
304
+ | Qwen3-235b | 0.02 | 0.00 | +0.02 |
305
+
306
+ ## any_order_match
307
+
308
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
309
+ | --- | --- | --- | --- |
310
+ | DeepSeek-R1 | 0.00 | 0.25 | -0.25 |
311
+ | DeepSeek-V3-1 | 0.00 | 1.00 | -1.00 |
312
+ | GPT-4o-mini | 0.68 | 0.72 | -0.03 |
313
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
314
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
315
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
316
+ | Qwen3-235b | 0.57 | 0.22 | +0.35 |
317
+
318
+ ## precision
319
+
320
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
321
+ | --- | --- | --- | --- |
322
+ | DeepSeek-R1 | 0.83 | 0.89 | -0.06 |
323
+ | DeepSeek-V3-1 | 0.87 | 0.82 | +0.05 |
324
+ | GPT-4o-mini | 0.79 | 0.75 | +0.05 |
325
+ | GPT-5 | 0.99 | 0.97 | +0.02 |
326
+ | Gemini-2.5-flash | 1.00 | 0.94 | +0.06 |
327
+ | Gemini-2.5-flash-nothinking | 1.00 | 0.93 | +0.06 |
328
+ | Qwen3-235b | 0.74 | 0.76 | -0.02 |
329
+
330
+ ## recall
331
+
332
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
333
+ | --- | --- | --- | --- |
334
+ | DeepSeek-R1 | 0.88 | 0.93 | -0.06 |
335
+ | DeepSeek-V3-1 | 0.96 | 1.00 | -0.04 |
336
+ | GPT-4o-mini | 0.98 | 0.99 | -0.00 |
337
+ | GPT-5 | 0.90 | 0.91 | -0.01 |
338
+ | Gemini-2.5-flash | 0.64 | 0.89 | -0.25 |
339
+ | Gemini-2.5-flash-nothinking | 0.68 | 0.94 | -0.26 |
340
+ | Qwen3-235b | 0.97 | 0.96 | +0.01 |
341
+
342
+ ## retry_rate
343
+
344
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
345
+ | --- | --- | --- | --- |
346
+ | DeepSeek-R1 | 0.07 | 0.05 | +0.02 |
347
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
348
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
349
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
350
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
351
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
352
+ | Qwen3-235b | 0.27 | 0.18 | +0.08 |
353
+
354
+ ## pass_rate
355
+
356
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
357
+ | --- | --- | --- | --- |
358
+ | DeepSeek-R1 | 0.98 | 1.00 | -0.02 |
359
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
360
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
361
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
362
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
363
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
364
+ | Qwen3-235b | 0.98 | 1.00 | -0.02 |
365
+
366
+ # SQLAssistant: A2A vs H_A2A
367
+
368
+ ## Overall Summary (Averaged Across All Models)
369
+
370
+ | Metric | A2A | H_A2A | Diff (A2A-H_A2A) |
371
+ | --- | --- | --- | --- |
372
+ | exact_match | 0.00 | 0.00 | +0.00 |
373
+ | any_order_match | 0.15 | 0.00 | +0.15 |
374
+ | precision | 0.65 | 0.83 | -0.18 |
375
+ | recall | 0.87 | 0.87 | -0.01 |
376
+ | retry_rate | 0.65 | 0.38 | +0.27 |
377
+ | average_score | 0.32 | 0.57 | -0.25 |
378
+
379
+ ---
380
+
381
+ ## exact_match
382
+
383
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
384
+ | --- | --- | --- | --- |
385
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
386
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
387
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
388
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
389
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
390
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
391
+ | Qwen3-235b | 0.00 | 0.00 | +0.00 |
392
+
393
+ ## any_order_match
394
+
395
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
396
+ | --- | --- | --- | --- |
397
+ | DeepSeek-R1 | 0.03 | 0.00 | +0.03 |
398
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
399
+ | GPT-4o-mini | 0.53 | 0.00 | +0.53 |
400
+ | GPT-5 | 0.03 | 0.00 | +0.03 |
401
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
402
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
403
+ | Qwen3-235b | 0.48 | 0.00 | +0.48 |
404
+
405
+ ## precision
406
+
407
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
408
+ | --- | --- | --- | --- |
409
+ | DeepSeek-R1 | 0.41 | 0.64 | -0.23 |
410
+ | DeepSeek-V3-1 | 0.41 | 0.89 | -0.48 |
411
+ | GPT-4o-mini | 0.84 | 0.86 | -0.02 |
412
+ | GPT-5 | 0.84 | 0.91 | -0.07 |
413
+ | Gemini-2.5-flash | 0.57 | 0.81 | -0.23 |
414
+ | Gemini-2.5-flash-nothinking | 0.70 | 0.84 | -0.14 |
415
+ | Qwen3-235b | 0.78 | 0.85 | -0.07 |
416
+
417
+ ## recall
418
+
419
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
420
+ | --- | --- | --- | --- |
421
+ | DeepSeek-R1 | 0.85 | 0.89 | -0.03 |
422
+ | DeepSeek-V3-1 | 0.84 | 0.90 | -0.05 |
423
+ | GPT-4o-mini | 0.98 | 0.92 | +0.07 |
424
+ | GPT-5 | 0.88 | 0.86 | +0.02 |
425
+ | Gemini-2.5-flash | 0.77 | 0.83 | -0.05 |
426
+ | Gemini-2.5-flash-nothinking | 0.78 | 0.80 | -0.02 |
427
+ | Qwen3-235b | 0.96 | 0.93 | +0.02 |
428
+
429
+ ## retry_rate
430
+
431
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
432
+ | --- | --- | --- | --- |
433
+ | DeepSeek-R1 | 0.95 | 0.67 | +0.28 |
434
+ | DeepSeek-V3-1 | 1.00 | 0.27 | +0.73 |
435
+ | GPT-4o-mini | 0.17 | 0.28 | -0.12 |
436
+ | GPT-5 | 0.40 | 0.23 | +0.17 |
437
+ | Gemini-2.5-flash | 0.90 | 0.50 | +0.40 |
438
+ | Gemini-2.5-flash-nothinking | 0.75 | 0.37 | +0.38 |
439
+ | Qwen3-235b | 0.38 | 0.33 | +0.05 |
440
+
441
+ ## average_score
442
+
443
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
444
+ | --- | --- | --- | --- |
445
+ | DeepSeek-R1 | 0.25 | 0.52 | -0.27 |
446
+ | DeepSeek-V3-1 | 0.12 | 0.53 | -0.41 |
447
+ | GPT-4o-mini | 0.58 | 0.58 | +0.00 |
448
+ | GPT-5 | 0.47 | 0.53 | -0.06 |
449
+ | Gemini-2.5-flash | 0.16 | 0.55 | -0.39 |
450
+ | Gemini-2.5-flash-nothinking | 0.05 | 0.57 | -0.52 |
451
+ | Qwen3-235b | 0.60 | 0.68 | -0.08 |
452
+
453
+ # SocialMediaManager: A2A vs H_A2A
454
+
455
+ ## Overall Summary (Averaged Across All Models)
456
+
457
+ | Metric | A2A | H_A2A | Diff (A2A-H_A2A) |
458
+ | --- | --- | --- | --- |
459
+ | exact_match | 0.04 | 0.07 | -0.04 |
460
+ | any_order_match | 0.42 | 0.29 | +0.13 |
461
+ | precision | 0.80 | 0.85 | -0.05 |
462
+ | recall | 0.79 | 0.87 | -0.08 |
463
+ | retry_rate | 0.39 | 0.29 | +0.09 |
464
+ | pass_rate | 0.97 | 0.99 | -0.02 |
465
+
466
+ ---
467
+
468
+ ## exact_match
469
+
470
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
471
+ | --- | --- | --- | --- |
472
+ | DeepSeek-R1 | 0.00 | 0.02 | -0.02 |
473
+ | DeepSeek-V3-1 | 0.15 | 0.20 | -0.05 |
474
+ | GPT-4o-mini | 0.03 | 0.27 | -0.23 |
475
+ | GPT-5 | 0.05 | 0.00 | +0.05 |
476
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
477
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
478
+ | Qwen3-235b | 0.03 | 0.03 | +0.00 |
479
+
480
+ ## any_order_match
481
+
482
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
483
+ | --- | --- | --- | --- |
484
+ | DeepSeek-R1 | 0.13 | 0.32 | -0.18 |
485
+ | DeepSeek-V3-1 | 0.85 | 0.83 | +0.02 |
486
+ | GPT-4o-mini | 0.80 | 0.45 | +0.35 |
487
+ | GPT-5 | 0.53 | 0.15 | +0.38 |
488
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
489
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
490
+ | Qwen3-235b | 0.62 | 0.27 | +0.35 |
491
+
492
+ ## precision
493
+
494
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
495
+ | --- | --- | --- | --- |
496
+ | DeepSeek-R1 | 0.60 | 0.70 | -0.10 |
497
+ | DeepSeek-V3-1 | 0.79 | 0.83 | -0.04 |
498
+ | GPT-4o-mini | 0.66 | 0.90 | -0.24 |
499
+ | GPT-5 | 0.83 | 0.92 | -0.09 |
500
+ | Gemini-2.5-flash | 0.98 | 0.87 | +0.11 |
501
+ | Gemini-2.5-flash-nothinking | 0.97 | 0.93 | +0.04 |
502
+ | Qwen3-235b | 0.80 | 0.82 | -0.02 |
503
+
504
+ ## recall
505
+
506
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
507
+ | --- | --- | --- | --- |
508
+ | DeepSeek-R1 | 0.89 | 0.90 | -0.01 |
509
+ | DeepSeek-V3-1 | 0.98 | 0.96 | +0.01 |
510
+ | GPT-4o-mini | 0.99 | 0.95 | +0.03 |
511
+ | GPT-5 | 0.90 | 0.87 | +0.02 |
512
+ | Gemini-2.5-flash | 0.44 | 0.77 | -0.33 |
513
+ | Gemini-2.5-flash-nothinking | 0.43 | 0.76 | -0.32 |
514
+ | Qwen3-235b | 0.93 | 0.88 | +0.05 |
515
+
516
+ ## retry_rate
517
+
518
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
519
+ | --- | --- | --- | --- |
520
+ | DeepSeek-R1 | 0.65 | 0.55 | +0.10 |
521
+ | DeepSeek-V3-1 | 0.23 | 0.18 | +0.05 |
522
+ | GPT-4o-mini | 0.60 | 0.15 | +0.45 |
523
+ | GPT-5 | 0.48 | 0.08 | +0.40 |
524
+ | Gemini-2.5-flash | 0.07 | 0.42 | -0.35 |
525
+ | Gemini-2.5-flash-nothinking | 0.12 | 0.25 | -0.13 |
526
+ | Qwen3-235b | 0.55 | 0.42 | +0.13 |
527
+
528
+ ## pass_rate
529
+
530
+ | Model | A2A | H_A2A | Diff (A2A-H_A2A) |
531
+ | --- | --- | --- | --- |
532
+ | DeepSeek-R1 | 0.92 | 0.97 | -0.05 |
533
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
534
+ | GPT-4o-mini | 0.93 | 1.00 | -0.07 |
535
+ | GPT-5 | 0.97 | 1.00 | -0.03 |
536
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
537
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
538
+ | Qwen3-235b | 0.98 | 0.98 | +0.00 |
539
+
data/processed/RQ1/RadarCharts/Comparison_MCP_vs_A2A.md CHANGED
@@ -1,11 +1,18 @@
1
  # Overall MCP vs A2A Comparison
2
 
3
- Averaged across all projects: SQL_assistant, intelligent_recruitment_platform, landing_page_generator, self_evaluation_loop_flow, write_a_book_with_flows
4
 
5
  ## Framework-Level Comparison (All Models Averaged)
6
 
7
  | Metric | MCP | A2A | Diff (MCP-A2A) |
8
  | --- | --- | --- | --- |
 
 
 
 
 
 
 
9
 
10
  ---
11
 
@@ -13,56 +20,520 @@ Averaged across all projects: SQL_assistant, intelligent_recruitment_platform, l
13
 
14
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
15
  | --- | --- | --- | --- |
 
 
 
 
 
 
 
16
 
17
  ## any_order_match
18
 
19
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
20
  | --- | --- | --- | --- |
 
 
 
 
 
 
 
21
 
22
  ## precision
23
 
24
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
25
  | --- | --- | --- | --- |
 
 
 
 
 
 
 
26
 
27
  ## recall
28
 
29
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
30
  | --- | --- | --- | --- |
 
 
 
 
 
 
 
31
 
32
  ## retry_rate
33
 
34
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
35
  | --- | --- | --- | --- |
 
 
 
 
 
 
 
36
 
37
  ## average_score
38
 
39
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
40
  | --- | --- | --- | --- |
 
 
 
 
 
 
 
41
 
42
  ## pass_rate
43
 
44
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
45
  | --- | --- | --- | --- |
 
 
 
 
 
 
 
46
 
47
  ---
48
 
49
- # SQL_assistant: MCP vs A2A
50
 
51
- _Data not available for comparison_
52
 
53
- # intelligent_recruitment_platform: MCP vs A2A
 
 
 
 
 
 
 
 
 
 
 
54
 
55
- _Data not available for comparison_
 
 
 
 
 
 
 
 
56
 
57
- # landing_page_generator: MCP vs A2A
58
 
59
- _Data not available for comparison_
 
 
 
 
 
 
 
 
60
 
61
- # self_evaluation_loop_flow: MCP vs A2A
62
 
63
- _Data not available for comparison_
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
64
 
65
- # write_a_book_with_flows: MCP vs A2A
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
66
 
67
- _Data not available for comparison_
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
68
 
 
1
  # Overall MCP vs A2A Comparison
2
 
3
+ Averaged across all projects: BookWriter, LandingPageGenerator, RecruitmentAssistant, SQLAssistant, SocialMediaManager
4
 
5
  ## Framework-Level Comparison (All Models Averaged)
6
 
7
  | Metric | MCP | A2A | Diff (MCP-A2A) |
8
  | --- | --- | --- | --- |
9
+ | exact_match | 0.07 | 0.01 | +0.06 |
10
+ | any_order_match | 0.29 | 0.25 | +0.04 |
11
+ | precision | 0.85 | 0.82 | +0.03 |
12
+ | recall | 0.81 | 0.84 | -0.03 |
13
+ | retry_rate | 0.21 | 0.22 | -0.02 |
14
+ | average_score | 0.47 | 0.31 | +0.17 |
15
+ | pass_rate | 0.99 | 0.99 | +0.00 |
16
 
17
  ---
18
 
 
20
 
21
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
22
  | --- | --- | --- | --- |
23
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
24
+ | DeepSeek-V3-1 | 0.20 | 0.03 | +0.17 |
25
+ | GPT-4o-mini | 0.04 | 0.01 | +0.03 |
26
+ | GPT-5 | 0.18 | 0.01 | +0.17 |
27
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
28
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
29
+ | Qwen3-235b | 0.07 | 0.01 | +0.06 |
30
 
31
  ## any_order_match
32
 
33
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
34
  | --- | --- | --- | --- |
35
+ | DeepSeek-R1 | 0.14 | 0.05 | +0.10 |
36
+ | DeepSeek-V3-1 | 0.63 | 0.48 | +0.15 |
37
+ | GPT-4o-mini | 0.54 | 0.60 | -0.06 |
38
+ | GPT-5 | 0.29 | 0.17 | +0.12 |
39
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
40
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
41
+ | Qwen3-235b | 0.42 | 0.43 | -0.01 |
42
 
43
  ## precision
44
 
45
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
46
  | --- | --- | --- | --- |
47
+ | DeepSeek-R1 | 0.71 | 0.70 | +0.02 |
48
+ | DeepSeek-V3-1 | 0.84 | 0.74 | +0.10 |
49
+ | GPT-4o-mini | 0.72 | 0.79 | -0.06 |
50
+ | GPT-5 | 0.90 | 0.89 | +0.02 |
51
+ | Gemini-2.5-flash | 0.97 | 0.91 | +0.06 |
52
+ | Gemini-2.5-flash-nothinking | 0.97 | 0.93 | +0.04 |
53
+ | Qwen3-235b | 0.84 | 0.81 | +0.03 |
54
 
55
  ## recall
56
 
57
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
58
  | --- | --- | --- | --- |
59
+ | DeepSeek-R1 | 0.88 | 0.88 | -0.00 |
60
+ | DeepSeek-V3-1 | 0.98 | 0.95 | +0.02 |
61
+ | GPT-4o-mini | 0.95 | 0.96 | -0.01 |
62
+ | GPT-5 | 0.88 | 0.90 | -0.02 |
63
+ | Gemini-2.5-flash | 0.55 | 0.62 | -0.07 |
64
+ | Gemini-2.5-flash-nothinking | 0.57 | 0.64 | -0.07 |
65
+ | Qwen3-235b | 0.90 | 0.93 | -0.03 |
66
 
67
  ## retry_rate
68
 
69
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
70
  | --- | --- | --- | --- |
71
+ | DeepSeek-R1 | 0.35 | 0.37 | -0.02 |
72
+ | DeepSeek-V3-1 | 0.19 | 0.25 | -0.06 |
73
+ | GPT-4o-mini | 0.26 | 0.15 | +0.11 |
74
+ | GPT-5 | 0.23 | 0.18 | +0.05 |
75
+ | Gemini-2.5-flash | 0.09 | 0.19 | -0.10 |
76
+ | Gemini-2.5-flash-nothinking | 0.09 | 0.18 | -0.09 |
77
+ | Qwen3-235b | 0.23 | 0.25 | -0.02 |
78
 
79
  ## average_score
80
 
81
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
82
  | --- | --- | --- | --- |
83
+ | DeepSeek-R1 | 0.51 | 0.16 | +0.34 |
84
+ | DeepSeek-V3-1 | 0.66 | 0.49 | +0.17 |
85
+ | GPT-4o-mini | 0.63 | 0.57 | +0.06 |
86
+ | GPT-5 | 0.63 | 0.33 | +0.30 |
87
+ | Gemini-2.5-flash | 0.09 | 0.08 | +0.01 |
88
+ | Gemini-2.5-flash-nothinking | 0.10 | 0.03 | +0.08 |
89
+ | Qwen3-235b | 0.70 | 0.49 | +0.21 |
90
 
91
  ## pass_rate
92
 
93
  | Model | MCP (Avg) | A2A (Avg) | Diff (MCP-A2A) |
94
  | --- | --- | --- | --- |
95
+ | DeepSeek-R1 | 0.98 | 0.97 | +0.02 |
96
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
97
+ | GPT-4o-mini | 0.98 | 0.98 | +0.01 |
98
+ | GPT-5 | 0.99 | 0.99 | +0.01 |
99
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
100
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
101
+ | Qwen3-235b | 0.98 | 0.99 | -0.01 |
102
 
103
  ---
104
 
105
+ # BookWriter: MCP vs A2A
106
 
107
+ ## Overall Summary (Averaged Across All Models)
108
 
109
+ | Metric | MCP | A2A | Diff (MCP-A2A) |
110
+ | --- | --- | --- | --- |
111
+ | exact_match | 0.02 | 0.00 | +0.02 |
112
+ | any_order_match | 0.09 | 0.08 | +0.01 |
113
+ | precision | 0.94 | 0.93 | +0.01 |
114
+ | recall | 0.73 | 0.77 | -0.04 |
115
+ | retry_rate | 0.02 | 0.02 | +0.00 |
116
+ | pass_rate | 1.00 | 1.00 | +0.00 |
117
+
118
+ ---
119
+
120
+ ## exact_match
121
 
122
+ | Model | MCP | A2A | Diff (MCP-A2A) |
123
+ | --- | --- | --- | --- |
124
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
125
+ | DeepSeek-V3-1 | 0.13 | 0.00 | +0.13 |
126
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
127
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
128
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
129
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
130
+ | Qwen3-235b | 0.00 | 0.00 | +0.00 |
131
 
132
+ ## any_order_match
133
 
134
+ | Model | MCP | A2A | Diff (MCP-A2A) |
135
+ | --- | --- | --- | --- |
136
+ | DeepSeek-R1 | 0.02 | 0.02 | +0.00 |
137
+ | DeepSeek-V3-1 | 0.63 | 0.53 | +0.10 |
138
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
139
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
140
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
141
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
142
+ | Qwen3-235b | 0.00 | 0.00 | +0.00 |
143
 
144
+ ## precision
145
 
146
+ | Model | MCP | A2A | Diff (MCP-A2A) |
147
+ | --- | --- | --- | --- |
148
+ | DeepSeek-R1 | 0.86 | 0.87 | -0.02 |
149
+ | DeepSeek-V3-1 | 0.96 | 0.79 | +0.17 |
150
+ | GPT-4o-mini | 0.84 | 0.90 | -0.06 |
151
+ | GPT-5 | 0.98 | 0.98 | +0.01 |
152
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
153
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | -0.00 |
154
+ | Qwen3-235b | 0.96 | 0.96 | -0.00 |
155
+
156
+ ## recall
157
+
158
+ | Model | MCP | A2A | Diff (MCP-A2A) |
159
+ | --- | --- | --- | --- |
160
+ | DeepSeek-R1 | 0.80 | 0.84 | -0.04 |
161
+ | DeepSeek-V3-1 | 0.99 | 0.99 | -0.00 |
162
+ | GPT-4o-mini | 0.85 | 0.86 | -0.01 |
163
+ | GPT-5 | 0.81 | 0.82 | -0.02 |
164
+ | Gemini-2.5-flash | 0.47 | 0.53 | -0.06 |
165
+ | Gemini-2.5-flash-nothinking | 0.47 | 0.53 | -0.06 |
166
+ | Qwen3-235b | 0.73 | 0.80 | -0.06 |
167
+
168
+ ## retry_rate
169
+
170
+ | Model | MCP | A2A | Diff (MCP-A2A) |
171
+ | --- | --- | --- | --- |
172
+ | DeepSeek-R1 | 0.13 | 0.13 | +0.00 |
173
+ | DeepSeek-V3-1 | 0.00 | 0.02 | -0.02 |
174
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
175
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
176
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
177
+ | Gemini-2.5-flash-nothinking | 0.02 | 0.00 | +0.02 |
178
+ | Qwen3-235b | 0.02 | 0.00 | +0.02 |
179
 
180
+ ## pass_rate
181
+
182
+ | Model | MCP | A2A | Diff (MCP-A2A) |
183
+ | --- | --- | --- | --- |
184
+ | DeepSeek-R1 | 1.00 | 1.00 | +0.00 |
185
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
186
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
187
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
188
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
189
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
190
+ | Qwen3-235b | 1.00 | 1.00 | +0.00 |
191
+
192
+ # LandingPageGenerator: MCP vs A2A
193
+
194
+ ## Overall Summary (Averaged Across All Models)
195
+
196
+ | Metric | MCP | A2A | Diff (MCP-A2A) |
197
+ | --- | --- | --- | --- |
198
+ | exact_match | 0.29 | 0.00 | +0.29 |
199
+ | any_order_match | 0.60 | 0.40 | +0.20 |
200
+ | precision | 0.92 | 0.84 | +0.08 |
201
+ | recall | 0.90 | 0.91 | -0.01 |
202
+ | retry_rate | 0.03 | 0.02 | +0.01 |
203
+ | average_score | 0.57 | 0.30 | +0.27 |
204
+
205
+ ---
206
+
207
+ ## exact_match
208
+
209
+ | Model | MCP | A2A | Diff (MCP-A2A) |
210
+ | --- | --- | --- | --- |
211
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
212
+ | DeepSeek-V3-1 | 0.72 | 0.00 | +0.72 |
213
+ | GPT-4o-mini | 0.12 | 0.00 | +0.12 |
214
+ | GPT-5 | 0.88 | 0.00 | +0.88 |
215
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
216
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
217
+ | Qwen3-235b | 0.28 | 0.00 | +0.28 |
218
+
219
+ ## any_order_match
220
+
221
+ | Model | MCP | A2A | Diff (MCP-A2A) |
222
+ | --- | --- | --- | --- |
223
+ | DeepSeek-R1 | 0.32 | 0.05 | +0.27 |
224
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
225
+ | GPT-4o-mini | 0.98 | 0.97 | +0.02 |
226
+ | GPT-5 | 1.00 | 0.27 | +0.73 |
227
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
228
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
229
+ | Qwen3-235b | 0.87 | 0.48 | +0.38 |
230
 
231
+ ## precision
232
+
233
+ | Model | MCP | A2A | Diff (MCP-A2A) |
234
+ | --- | --- | --- | --- |
235
+ | DeepSeek-R1 | 0.80 | 0.76 | +0.04 |
236
+ | DeepSeek-V3-1 | 0.97 | 0.83 | +0.13 |
237
+ | GPT-4o-mini | 0.82 | 0.74 | +0.08 |
238
+ | GPT-5 | 0.99 | 0.81 | +0.18 |
239
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
240
+ | Gemini-2.5-flash-nothinking | 0.99 | 0.98 | +0.00 |
241
+ | Qwen3-235b | 0.91 | 0.79 | +0.12 |
242
+
243
+ ## recall
244
+
245
+ | Model | MCP | A2A | Diff (MCP-A2A) |
246
+ | --- | --- | --- | --- |
247
+ | DeepSeek-R1 | 0.94 | 0.94 | +0.00 |
248
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
249
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
250
+ | GPT-5 | 1.00 | 0.97 | +0.03 |
251
+ | Gemini-2.5-flash | 0.67 | 0.74 | -0.07 |
252
+ | Gemini-2.5-flash-nothinking | 0.70 | 0.76 | -0.07 |
253
+ | Qwen3-235b | 0.99 | 0.98 | +0.01 |
254
+
255
+ ## retry_rate
256
+
257
+ | Model | MCP | A2A | Diff (MCP-A2A) |
258
+ | --- | --- | --- | --- |
259
+ | DeepSeek-R1 | 0.15 | 0.03 | +0.12 |
260
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
261
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
262
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
263
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
264
+ | Gemini-2.5-flash-nothinking | 0.03 | 0.05 | -0.02 |
265
+ | Qwen3-235b | 0.00 | 0.05 | -0.05 |
266
+
267
+ ## average_score
268
+
269
+ | Model | MCP | A2A | Diff (MCP-A2A) |
270
+ | --- | --- | --- | --- |
271
+ | DeepSeek-R1 | 0.58 | 0.08 | +0.50 |
272
+ | DeepSeek-V3-1 | 0.83 | 0.86 | -0.02 |
273
+ | GPT-4o-mini | 0.86 | 0.57 | +0.30 |
274
+ | GPT-5 | 0.91 | 0.19 | +0.71 |
275
+ | Gemini-2.5-flash | 0.01 | 0.00 | +0.01 |
276
+ | Gemini-2.5-flash-nothinking | 0.03 | 0.00 | +0.03 |
277
+ | Qwen3-235b | 0.78 | 0.38 | +0.40 |
278
+
279
+ # RecruitmentAssistant: MCP vs A2A
280
+
281
+ ## Overall Summary (Averaged Across All Models)
282
+
283
+ | Metric | MCP | A2A | Diff (MCP-A2A) |
284
+ | --- | --- | --- | --- |
285
+ | exact_match | 0.01 | 0.00 | +0.00 |
286
+ | any_order_match | 0.21 | 0.18 | +0.03 |
287
+ | precision | 0.87 | 0.89 | -0.02 |
288
+ | recall | 0.83 | 0.86 | -0.03 |
289
+ | retry_rate | 0.05 | 0.05 | +0.00 |
290
+ | pass_rate | 1.00 | 1.00 | +0.00 |
291
+
292
+ ---
293
+
294
+ ## exact_match
295
+
296
+ | Model | MCP | A2A | Diff (MCP-A2A) |
297
+ | --- | --- | --- | --- |
298
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
299
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
300
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
301
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
302
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
303
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
304
+ | Qwen3-235b | 0.05 | 0.02 | +0.03 |
305
+
306
+ ## any_order_match
307
+
308
+ | Model | MCP | A2A | Diff (MCP-A2A) |
309
+ | --- | --- | --- | --- |
310
+ | DeepSeek-R1 | 0.08 | 0.00 | +0.08 |
311
+ | DeepSeek-V3-1 | 0.07 | 0.00 | +0.07 |
312
+ | GPT-4o-mini | 0.70 | 0.68 | +0.02 |
313
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
314
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
315
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
316
+ | Qwen3-235b | 0.60 | 0.57 | +0.03 |
317
+
318
+ ## precision
319
+
320
+ | Model | MCP | A2A | Diff (MCP-A2A) |
321
+ | --- | --- | --- | --- |
322
+ | DeepSeek-R1 | 0.79 | 0.83 | -0.04 |
323
+ | DeepSeek-V3-1 | 0.84 | 0.87 | -0.03 |
324
+ | GPT-4o-mini | 0.74 | 0.79 | -0.05 |
325
+ | GPT-5 | 0.99 | 0.99 | -0.00 |
326
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
327
+ | Gemini-2.5-flash-nothinking | 0.99 | 1.00 | -0.00 |
328
+ | Qwen3-235b | 0.73 | 0.74 | -0.01 |
329
+
330
+ ## recall
331
+
332
+ | Model | MCP | A2A | Diff (MCP-A2A) |
333
+ | --- | --- | --- | --- |
334
+ | DeepSeek-R1 | 0.88 | 0.88 | +0.00 |
335
+ | DeepSeek-V3-1 | 0.94 | 0.96 | -0.02 |
336
+ | GPT-4o-mini | 0.98 | 0.98 | -0.01 |
337
+ | GPT-5 | 0.89 | 0.90 | -0.01 |
338
+ | Gemini-2.5-flash | 0.55 | 0.64 | -0.09 |
339
+ | Gemini-2.5-flash-nothinking | 0.60 | 0.68 | -0.08 |
340
+ | Qwen3-235b | 0.96 | 0.97 | -0.02 |
341
+
342
+ ## retry_rate
343
+
344
+ | Model | MCP | A2A | Diff (MCP-A2A) |
345
+ | --- | --- | --- | --- |
346
+ | DeepSeek-R1 | 0.03 | 0.07 | -0.03 |
347
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
348
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
349
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
350
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
351
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
352
+ | Qwen3-235b | 0.33 | 0.27 | +0.07 |
353
+
354
+ ## pass_rate
355
+
356
+ | Model | MCP | A2A | Diff (MCP-A2A) |
357
+ | --- | --- | --- | --- |
358
+ | DeepSeek-R1 | 1.00 | 0.98 | +0.02 |
359
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
360
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
361
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
362
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
363
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
364
+ | Qwen3-235b | 0.97 | 0.98 | -0.02 |
365
+
366
+ # SQLAssistant: MCP vs A2A
367
+
368
+ ## Overall Summary (Averaged Across All Models)
369
+
370
+ | Metric | MCP | A2A | Diff (MCP-A2A) |
371
+ | --- | --- | --- | --- |
372
+ | exact_match | 0.00 | 0.00 | +0.00 |
373
+ | any_order_match | 0.14 | 0.15 | -0.02 |
374
+ | precision | 0.72 | 0.65 | +0.07 |
375
+ | recall | 0.86 | 0.87 | -0.01 |
376
+ | retry_rate | 0.53 | 0.65 | -0.12 |
377
+ | average_score | 0.38 | 0.32 | +0.06 |
378
+
379
+ ---
380
+
381
+ ## exact_match
382
+
383
+ | Model | MCP | A2A | Diff (MCP-A2A) |
384
+ | --- | --- | --- | --- |
385
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
386
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
387
+ | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
388
+ | GPT-5 | 0.00 | 0.00 | +0.00 |
389
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
390
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
391
+ | Qwen3-235b | 0.00 | 0.00 | +0.00 |
392
+
393
+ ## any_order_match
394
+
395
+ | Model | MCP | A2A | Diff (MCP-A2A) |
396
+ | --- | --- | --- | --- |
397
+ | DeepSeek-R1 | 0.10 | 0.03 | +0.07 |
398
+ | DeepSeek-V3-1 | 0.52 | 0.00 | +0.52 |
399
+ | GPT-4o-mini | 0.12 | 0.53 | -0.42 |
400
+ | GPT-5 | 0.00 | 0.03 | -0.03 |
401
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
402
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
403
+ | Qwen3-235b | 0.23 | 0.48 | -0.25 |
404
+
405
+ ## precision
406
+
407
+ | Model | MCP | A2A | Diff (MCP-A2A) |
408
+ | --- | --- | --- | --- |
409
+ | DeepSeek-R1 | 0.50 | 0.41 | +0.10 |
410
+ | DeepSeek-V3-1 | 0.67 | 0.41 | +0.26 |
411
+ | GPT-4o-mini | 0.61 | 0.84 | -0.23 |
412
+ | GPT-5 | 0.71 | 0.84 | -0.13 |
413
+ | Gemini-2.5-flash | 0.89 | 0.57 | +0.32 |
414
+ | Gemini-2.5-flash-nothinking | 0.90 | 0.70 | +0.20 |
415
+ | Qwen3-235b | 0.78 | 0.78 | -0.00 |
416
+
417
+ ## recall
418
+
419
+ | Model | MCP | A2A | Diff (MCP-A2A) |
420
+ | --- | --- | --- | --- |
421
+ | DeepSeek-R1 | 0.90 | 0.85 | +0.05 |
422
+ | DeepSeek-V3-1 | 0.96 | 0.84 | +0.12 |
423
+ | GPT-4o-mini | 0.92 | 0.98 | -0.06 |
424
+ | GPT-5 | 0.85 | 0.88 | -0.04 |
425
+ | Gemini-2.5-flash | 0.71 | 0.77 | -0.07 |
426
+ | Gemini-2.5-flash-nothinking | 0.71 | 0.78 | -0.07 |
427
+ | Qwen3-235b | 0.96 | 0.96 | -0.00 |
428
+
429
+ ## retry_rate
430
+
431
+ | Model | MCP | A2A | Diff (MCP-A2A) |
432
+ | --- | --- | --- | --- |
433
+ | DeepSeek-R1 | 0.78 | 0.95 | -0.17 |
434
+ | DeepSeek-V3-1 | 0.57 | 1.00 | -0.43 |
435
+ | GPT-4o-mini | 0.65 | 0.17 | +0.48 |
436
+ | GPT-5 | 0.68 | 0.40 | +0.28 |
437
+ | Gemini-2.5-flash | 0.30 | 0.90 | -0.60 |
438
+ | Gemini-2.5-flash-nothinking | 0.32 | 0.75 | -0.43 |
439
+ | Qwen3-235b | 0.38 | 0.38 | +0.00 |
440
+
441
+ ## average_score
442
+
443
+ | Model | MCP | A2A | Diff (MCP-A2A) |
444
+ | --- | --- | --- | --- |
445
+ | DeepSeek-R1 | 0.44 | 0.25 | +0.19 |
446
+ | DeepSeek-V3-1 | 0.48 | 0.12 | +0.36 |
447
+ | GPT-4o-mini | 0.40 | 0.58 | -0.18 |
448
+ | GPT-5 | 0.36 | 0.47 | -0.11 |
449
+ | Gemini-2.5-flash | 0.16 | 0.16 | +0.01 |
450
+ | Gemini-2.5-flash-nothinking | 0.18 | 0.05 | +0.13 |
451
+ | Qwen3-235b | 0.62 | 0.60 | +0.02 |
452
+
453
+ # SocialMediaManager: MCP vs A2A
454
+
455
+ ## Overall Summary (Averaged Across All Models)
456
+
457
+ | Metric | MCP | A2A | Diff (MCP-A2A) |
458
+ | --- | --- | --- | --- |
459
+ | exact_match | 0.03 | 0.04 | -0.01 |
460
+ | any_order_match | 0.40 | 0.42 | -0.01 |
461
+ | precision | 0.80 | 0.80 | -0.00 |
462
+ | recall | 0.75 | 0.79 | -0.04 |
463
+ | retry_rate | 0.40 | 0.39 | +0.01 |
464
+ | pass_rate | 0.98 | 0.97 | +0.01 |
465
+
466
+ ---
467
+
468
+ ## exact_match
469
+
470
+ | Model | MCP | A2A | Diff (MCP-A2A) |
471
+ | --- | --- | --- | --- |
472
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
473
+ | DeepSeek-V3-1 | 0.15 | 0.15 | +0.00 |
474
+ | GPT-4o-mini | 0.07 | 0.03 | +0.03 |
475
+ | GPT-5 | 0.00 | 0.05 | -0.05 |
476
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
477
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
478
+ | Qwen3-235b | 0.00 | 0.03 | -0.03 |
479
+
480
+ ## any_order_match
481
+
482
+ | Model | MCP | A2A | Diff (MCP-A2A) |
483
+ | --- | --- | --- | --- |
484
+ | DeepSeek-R1 | 0.20 | 0.13 | +0.07 |
485
+ | DeepSeek-V3-1 | 0.93 | 0.85 | +0.08 |
486
+ | GPT-4o-mini | 0.88 | 0.80 | +0.08 |
487
+ | GPT-5 | 0.43 | 0.53 | -0.10 |
488
+ | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
489
+ | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
490
+ | Qwen3-235b | 0.38 | 0.62 | -0.23 |
491
+
492
+ ## precision
493
+
494
+ | Model | MCP | A2A | Diff (MCP-A2A) |
495
+ | --- | --- | --- | --- |
496
+ | DeepSeek-R1 | 0.62 | 0.60 | +0.02 |
497
+ | DeepSeek-V3-1 | 0.74 | 0.79 | -0.05 |
498
+ | GPT-4o-mini | 0.61 | 0.66 | -0.04 |
499
+ | GPT-5 | 0.85 | 0.83 | +0.02 |
500
+ | Gemini-2.5-flash | 0.97 | 0.98 | -0.02 |
501
+ | Gemini-2.5-flash-nothinking | 0.98 | 0.97 | +0.01 |
502
+ | Qwen3-235b | 0.85 | 0.80 | +0.05 |
503
+
504
+ ## recall
505
+
506
+ | Model | MCP | A2A | Diff (MCP-A2A) |
507
+ | --- | --- | --- | --- |
508
+ | DeepSeek-R1 | 0.86 | 0.89 | -0.02 |
509
+ | DeepSeek-V3-1 | 0.98 | 0.98 | +0.01 |
510
+ | GPT-4o-mini | 0.99 | 0.99 | +0.00 |
511
+ | GPT-5 | 0.86 | 0.90 | -0.04 |
512
+ | Gemini-2.5-flash | 0.35 | 0.44 | -0.09 |
513
+ | Gemini-2.5-flash-nothinking | 0.35 | 0.43 | -0.09 |
514
+ | Qwen3-235b | 0.87 | 0.93 | -0.06 |
515
+
516
+ ## retry_rate
517
+
518
+ | Model | MCP | A2A | Diff (MCP-A2A) |
519
+ | --- | --- | --- | --- |
520
+ | DeepSeek-R1 | 0.65 | 0.65 | +0.00 |
521
+ | DeepSeek-V3-1 | 0.38 | 0.23 | +0.15 |
522
+ | GPT-4o-mini | 0.67 | 0.60 | +0.07 |
523
+ | GPT-5 | 0.45 | 0.48 | -0.03 |
524
+ | Gemini-2.5-flash | 0.15 | 0.07 | +0.08 |
525
+ | Gemini-2.5-flash-nothinking | 0.08 | 0.12 | -0.03 |
526
+ | Qwen3-235b | 0.42 | 0.55 | -0.13 |
527
+
528
+ ## pass_rate
529
+
530
+ | Model | MCP | A2A | Diff (MCP-A2A) |
531
+ | --- | --- | --- | --- |
532
+ | DeepSeek-R1 | 0.95 | 0.92 | +0.03 |
533
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
534
+ | GPT-4o-mini | 0.95 | 0.93 | +0.02 |
535
+ | GPT-5 | 0.98 | 0.97 | +0.02 |
536
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
537
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
538
+ | Qwen3-235b | 0.98 | 0.98 | +0.00 |
539
 
data/processed/RQ1/RadarCharts/Comparison_MCP_vs_Hardcoded.md CHANGED
@@ -1,6 +1,6 @@
1
  # Overall MCP vs Hardcoded Comparison
2
 
3
- Averaged across all projects: MarkdownValidator, GameBuilder, EmailResponder
4
 
5
  ## Framework-Level Comparison (All Models Averaged)
6
 
@@ -12,7 +12,7 @@ Averaged across all projects: MarkdownValidator, GameBuilder, EmailResponder
12
  | recall | 0.91 | 0.91 | +0.00 |
13
  | retry_rate | 0.03 | 0.04 | -0.02 |
14
  | average_score | 0.60 | 0.60 | +0.00 |
15
- | pass_rate | 0.00 | 0.00 | +0.00 |
16
 
17
  ---
18
 
@@ -92,28 +92,28 @@ Averaged across all projects: MarkdownValidator, GameBuilder, EmailResponder
92
 
93
  | Model | MCP (Avg) | Hardcoded (Avg) | Diff (MCP-Hardcoded) |
94
  | --- | --- | --- | --- |
95
- | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
96
- | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
97
- | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
98
- | GPT-5 | 0.00 | 0.00 | +0.00 |
99
- | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
100
- | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
101
- | Qwen3-235b | 0.00 | 0.00 | +0.00 |
102
 
103
  ---
104
 
105
- # MarkdownValidator: MCP vs Hardcoded
106
 
107
  ## Overall Summary (Averaged Across All Models)
108
 
109
  | Metric | MCP | Hardcoded | Diff (MCP-Hardcoded) |
110
  | --- | --- | --- | --- |
111
- | exact_match | 0.75 | 0.77 | -0.02 |
112
- | any_order_match | 0.76 | 0.79 | -0.03 |
113
- | precision | 0.98 | 0.98 | -0.00 |
114
- | recall | 0.94 | 0.95 | -0.00 |
115
- | retry_rate | 0.04 | 0.04 | +0.00 |
116
- | average_score | 0.60 | 0.60 | +0.00 |
117
 
118
  ---
119
 
@@ -121,73 +121,73 @@ Averaged across all projects: MarkdownValidator, GameBuilder, EmailResponder
121
 
122
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
123
  | --- | --- | --- | --- |
124
- | DeepSeek-R1 | 0.97 | 1.00 | -0.03 |
125
- | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
126
- | GPT-4o-mini | 0.98 | 0.97 | +0.02 |
127
- | GPT-5 | 1.00 | 1.00 | +0.00 |
128
- | Gemini-2.5-flash | 0.20 | 0.32 | -0.12 |
129
- | Gemini-2.5-flash-nothinking | 0.10 | 0.17 | -0.07 |
130
- | Qwen3-235b | 1.00 | 0.97 | +0.03 |
131
 
132
  ## any_order_match
133
 
134
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
135
  | --- | --- | --- | --- |
136
- | DeepSeek-R1 | 1.00 | 1.00 | +0.00 |
137
- | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
138
- | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
139
- | GPT-5 | 1.00 | 1.00 | +0.00 |
140
- | Gemini-2.5-flash | 0.25 | 0.40 | -0.15 |
141
- | Gemini-2.5-flash-nothinking | 0.10 | 0.17 | -0.07 |
142
- | Qwen3-235b | 1.00 | 0.97 | +0.03 |
143
 
144
  ## precision
145
 
146
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
147
  | --- | --- | --- | --- |
148
- | DeepSeek-R1 | 0.99 | 1.00 | -0.01 |
149
- | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
150
- | GPT-4o-mini | 1.00 | 0.99 | +0.00 |
151
  | GPT-5 | 1.00 | 1.00 | +0.00 |
152
- | Gemini-2.5-flash | 0.88 | 0.86 | +0.02 |
153
- | Gemini-2.5-flash-nothinking | 0.95 | 0.97 | -0.02 |
154
- | Qwen3-235b | 1.00 | 1.00 | +0.00 |
155
 
156
  ## recall
157
 
158
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
159
  | --- | --- | --- | --- |
160
- | DeepSeek-R1 | 1.00 | 1.00 | +0.00 |
161
- | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
162
- | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
163
- | GPT-5 | 1.00 | 1.00 | +0.00 |
164
- | Gemini-2.5-flash | 0.86 | 0.89 | -0.03 |
165
- | Gemini-2.5-flash-nothinking | 0.73 | 0.74 | -0.01 |
166
- | Qwen3-235b | 1.00 | 0.99 | +0.01 |
167
 
168
  ## retry_rate
169
 
170
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
171
  | --- | --- | --- | --- |
172
- | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
173
- | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
174
  | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
175
  | GPT-5 | 0.00 | 0.00 | +0.00 |
176
- | Gemini-2.5-flash | 0.20 | 0.22 | -0.02 |
177
- | Gemini-2.5-flash-nothinking | 0.07 | 0.05 | +0.02 |
178
  | Qwen3-235b | 0.00 | 0.00 | +0.00 |
179
 
180
- ## average_score
181
 
182
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
183
  | --- | --- | --- | --- |
184
- | DeepSeek-R1 | 0.79 | 0.82 | -0.03 |
185
- | DeepSeek-V3-1 | 0.84 | 0.85 | -0.01 |
186
- | GPT-4o-mini | 0.84 | 0.81 | +0.03 |
187
- | GPT-5 | 0.70 | 0.72 | -0.03 |
188
- | Gemini-2.5-flash | 0.06 | 0.11 | -0.05 |
189
- | Gemini-2.5-flash-nothinking | 0.18 | 0.10 | +0.08 |
190
- | Qwen3-235b | 0.83 | 0.78 | +0.04 |
191
 
192
  # GameBuilder: MCP vs Hardcoded
193
 
@@ -200,7 +200,7 @@ Averaged across all projects: MarkdownValidator, GameBuilder, EmailResponder
200
  | precision | 0.98 | 0.99 | -0.01 |
201
  | recall | 0.94 | 0.93 | +0.00 |
202
  | retry_rate | 0.00 | 0.01 | -0.01 |
203
- | pass_rate | 0.00 | 0.00 | +0.00 |
204
 
205
  ---
206
 
@@ -268,26 +268,26 @@ Averaged across all projects: MarkdownValidator, GameBuilder, EmailResponder
268
 
269
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
270
  | --- | --- | --- | --- |
271
- | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
272
- | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
273
- | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
274
- | GPT-5 | 0.00 | 0.00 | +0.00 |
275
- | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
276
- | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
277
- | Qwen3-235b | 0.00 | 0.00 | +0.00 |
278
 
279
- # EmailResponder: MCP vs Hardcoded
280
 
281
  ## Overall Summary (Averaged Across All Models)
282
 
283
  | Metric | MCP | Hardcoded | Diff (MCP-Hardcoded) |
284
  | --- | --- | --- | --- |
285
- | exact_match | 0.10 | 0.10 | -0.01 |
286
- | any_order_match | 0.14 | 0.11 | +0.02 |
287
- | precision | 0.96 | 0.95 | +0.01 |
288
- | recall | 0.85 | 0.84 | +0.01 |
289
- | retry_rate | 0.04 | 0.07 | -0.04 |
290
- | pass_rate | 0.00 | 0.00 | +0.00 |
291
 
292
  ---
293
 
@@ -295,71 +295,71 @@ Averaged across all projects: MarkdownValidator, GameBuilder, EmailResponder
295
 
296
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
297
  | --- | --- | --- | --- |
298
- | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
299
- | DeepSeek-V3-1 | 0.06 | 0.26 | -0.20 |
300
- | GPT-4o-mini | 0.56 | 0.30 | +0.26 |
301
- | GPT-5 | 0.00 | 0.01 | -0.01 |
302
- | Gemini-2.5-flash | 0.00 | 0.01 | -0.01 |
303
- | Gemini-2.5-flash-nothinking | 0.02 | 0.07 | -0.05 |
304
- | Qwen3-235b | 0.04 | 0.08 | -0.04 |
305
 
306
  ## any_order_match
307
 
308
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
309
  | --- | --- | --- | --- |
310
- | DeepSeek-R1 | 0.03 | 0.02 | +0.01 |
311
- | DeepSeek-V3-1 | 0.22 | 0.26 | -0.04 |
312
- | GPT-4o-mini | 0.61 | 0.34 | +0.27 |
313
- | GPT-5 | 0.00 | 0.01 | -0.01 |
314
- | Gemini-2.5-flash | 0.00 | 0.01 | -0.01 |
315
- | Gemini-2.5-flash-nothinking | 0.05 | 0.07 | -0.02 |
316
- | Qwen3-235b | 0.05 | 0.09 | -0.04 |
317
 
318
  ## precision
319
 
320
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
321
  | --- | --- | --- | --- |
322
- | DeepSeek-R1 | 0.80 | 0.74 | +0.06 |
323
- | DeepSeek-V3-1 | 0.97 | 0.99 | -0.03 |
324
- | GPT-4o-mini | 0.99 | 0.99 | +0.00 |
325
  | GPT-5 | 1.00 | 1.00 | +0.00 |
326
- | Gemini-2.5-flash | 0.97 | 0.97 | +0.01 |
327
- | Gemini-2.5-flash-nothinking | 0.96 | 0.96 | +0.00 |
328
- | Qwen3-235b | 1.00 | 1.00 | -0.00 |
329
 
330
  ## recall
331
 
332
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
333
  | --- | --- | --- | --- |
334
- | DeepSeek-R1 | 0.90 | 0.89 | +0.00 |
335
- | DeepSeek-V3-1 | 0.88 | 0.88 | -0.00 |
336
- | GPT-4o-mini | 0.94 | 0.89 | +0.04 |
337
- | GPT-5 | 0.83 | 0.83 | -0.00 |
338
- | Gemini-2.5-flash | 0.75 | 0.72 | +0.04 |
339
- | Gemini-2.5-flash-nothinking | 0.80 | 0.82 | -0.01 |
340
- | Qwen3-235b | 0.84 | 0.85 | -0.00 |
341
 
342
  ## retry_rate
343
 
344
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
345
  | --- | --- | --- | --- |
346
- | DeepSeek-R1 | 0.04 | 0.28 | -0.24 |
347
- | DeepSeek-V3-1 | 0.02 | 0.02 | +0.00 |
348
  | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
349
  | GPT-5 | 0.00 | 0.00 | +0.00 |
350
- | Gemini-2.5-flash | 0.09 | 0.10 | -0.01 |
351
- | Gemini-2.5-flash-nothinking | 0.12 | 0.12 | +0.00 |
352
  | Qwen3-235b | 0.00 | 0.00 | +0.00 |
353
 
354
- ## pass_rate
355
 
356
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
357
  | --- | --- | --- | --- |
358
- | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
359
- | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
360
- | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
361
- | GPT-5 | 0.00 | 0.00 | +0.00 |
362
- | Gemini-2.5-flash | 0.00 | 0.00 | +0.00 |
363
- | Gemini-2.5-flash-nothinking | 0.00 | 0.00 | +0.00 |
364
- | Qwen3-235b | 0.00 | 0.00 | +0.00 |
365
 
 
1
  # Overall MCP vs Hardcoded Comparison
2
 
3
+ Averaged across all projects: EmailResponder, GameBuilder, MarkdownValidator
4
 
5
  ## Framework-Level Comparison (All Models Averaged)
6
 
 
12
  | recall | 0.91 | 0.91 | +0.00 |
13
  | retry_rate | 0.03 | 0.04 | -0.02 |
14
  | average_score | 0.60 | 0.60 | +0.00 |
15
+ | pass_rate | 1.00 | 0.99 | +0.01 |
16
 
17
  ---
18
 
 
92
 
93
  | Model | MCP (Avg) | Hardcoded (Avg) | Diff (MCP-Hardcoded) |
94
  | --- | --- | --- | --- |
95
+ | DeepSeek-R1 | 1.00 | 0.99 | +0.01 |
96
+ | DeepSeek-V3-1 | 1.00 | 0.98 | +0.02 |
97
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
98
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
99
+ | Gemini-2.5-flash | 1.00 | 0.99 | +0.01 |
100
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
101
+ | Qwen3-235b | 1.00 | 1.00 | +0.00 |
102
 
103
  ---
104
 
105
+ # EmailResponder: MCP vs Hardcoded
106
 
107
  ## Overall Summary (Averaged Across All Models)
108
 
109
  | Metric | MCP | Hardcoded | Diff (MCP-Hardcoded) |
110
  | --- | --- | --- | --- |
111
+ | exact_match | 0.10 | 0.10 | -0.01 |
112
+ | any_order_match | 0.14 | 0.11 | +0.02 |
113
+ | precision | 0.96 | 0.95 | +0.01 |
114
+ | recall | 0.85 | 0.84 | +0.01 |
115
+ | retry_rate | 0.04 | 0.07 | -0.04 |
116
+ | pass_rate | 1.00 | 1.00 | +0.00 |
117
 
118
  ---
119
 
 
121
 
122
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
123
  | --- | --- | --- | --- |
124
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
125
+ | DeepSeek-V3-1 | 0.06 | 0.26 | -0.20 |
126
+ | GPT-4o-mini | 0.56 | 0.30 | +0.26 |
127
+ | GPT-5 | 0.00 | 0.01 | -0.01 |
128
+ | Gemini-2.5-flash | 0.00 | 0.01 | -0.01 |
129
+ | Gemini-2.5-flash-nothinking | 0.02 | 0.07 | -0.05 |
130
+ | Qwen3-235b | 0.04 | 0.08 | -0.04 |
131
 
132
  ## any_order_match
133
 
134
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
135
  | --- | --- | --- | --- |
136
+ | DeepSeek-R1 | 0.03 | 0.02 | +0.01 |
137
+ | DeepSeek-V3-1 | 0.22 | 0.26 | -0.04 |
138
+ | GPT-4o-mini | 0.61 | 0.34 | +0.27 |
139
+ | GPT-5 | 0.00 | 0.01 | -0.01 |
140
+ | Gemini-2.5-flash | 0.00 | 0.01 | -0.01 |
141
+ | Gemini-2.5-flash-nothinking | 0.05 | 0.07 | -0.02 |
142
+ | Qwen3-235b | 0.05 | 0.09 | -0.04 |
143
 
144
  ## precision
145
 
146
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
147
  | --- | --- | --- | --- |
148
+ | DeepSeek-R1 | 0.80 | 0.74 | +0.06 |
149
+ | DeepSeek-V3-1 | 0.97 | 0.99 | -0.03 |
150
+ | GPT-4o-mini | 0.99 | 0.99 | +0.00 |
151
  | GPT-5 | 1.00 | 1.00 | +0.00 |
152
+ | Gemini-2.5-flash | 0.97 | 0.97 | +0.01 |
153
+ | Gemini-2.5-flash-nothinking | 0.96 | 0.96 | +0.00 |
154
+ | Qwen3-235b | 1.00 | 1.00 | -0.00 |
155
 
156
  ## recall
157
 
158
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
159
  | --- | --- | --- | --- |
160
+ | DeepSeek-R1 | 0.90 | 0.89 | +0.00 |
161
+ | DeepSeek-V3-1 | 0.88 | 0.88 | -0.00 |
162
+ | GPT-4o-mini | 0.94 | 0.89 | +0.04 |
163
+ | GPT-5 | 0.83 | 0.83 | -0.00 |
164
+ | Gemini-2.5-flash | 0.75 | 0.72 | +0.04 |
165
+ | Gemini-2.5-flash-nothinking | 0.80 | 0.82 | -0.01 |
166
+ | Qwen3-235b | 0.84 | 0.85 | -0.00 |
167
 
168
  ## retry_rate
169
 
170
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
171
  | --- | --- | --- | --- |
172
+ | DeepSeek-R1 | 0.04 | 0.28 | -0.24 |
173
+ | DeepSeek-V3-1 | 0.02 | 0.02 | +0.00 |
174
  | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
175
  | GPT-5 | 0.00 | 0.00 | +0.00 |
176
+ | Gemini-2.5-flash | 0.09 | 0.10 | -0.01 |
177
+ | Gemini-2.5-flash-nothinking | 0.12 | 0.12 | +0.00 |
178
  | Qwen3-235b | 0.00 | 0.00 | +0.00 |
179
 
180
+ ## pass_rate
181
 
182
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
183
  | --- | --- | --- | --- |
184
+ | DeepSeek-R1 | 1.00 | 0.98 | +0.02 |
185
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
186
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
187
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
188
+ | Gemini-2.5-flash | 1.00 | 0.99 | +0.01 |
189
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
190
+ | Qwen3-235b | 1.00 | 1.00 | +0.00 |
191
 
192
  # GameBuilder: MCP vs Hardcoded
193
 
 
200
  | precision | 0.98 | 0.99 | -0.01 |
201
  | recall | 0.94 | 0.93 | +0.00 |
202
  | retry_rate | 0.00 | 0.01 | -0.01 |
203
+ | pass_rate | 1.00 | 0.99 | +0.01 |
204
 
205
  ---
206
 
 
268
 
269
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
270
  | --- | --- | --- | --- |
271
+ | DeepSeek-R1 | 1.00 | 1.00 | +0.00 |
272
+ | DeepSeek-V3-1 | 1.00 | 0.96 | +0.04 |
273
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
274
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
275
+ | Gemini-2.5-flash | 1.00 | 1.00 | +0.00 |
276
+ | Gemini-2.5-flash-nothinking | 1.00 | 1.00 | +0.00 |
277
+ | Qwen3-235b | 1.00 | 1.00 | +0.00 |
278
 
279
+ # MarkdownValidator: MCP vs Hardcoded
280
 
281
  ## Overall Summary (Averaged Across All Models)
282
 
283
  | Metric | MCP | Hardcoded | Diff (MCP-Hardcoded) |
284
  | --- | --- | --- | --- |
285
+ | exact_match | 0.75 | 0.77 | -0.02 |
286
+ | any_order_match | 0.76 | 0.79 | -0.03 |
287
+ | precision | 0.98 | 0.98 | -0.00 |
288
+ | recall | 0.94 | 0.95 | -0.00 |
289
+ | retry_rate | 0.04 | 0.04 | +0.00 |
290
+ | average_score | 0.60 | 0.60 | +0.00 |
291
 
292
  ---
293
 
 
295
 
296
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
297
  | --- | --- | --- | --- |
298
+ | DeepSeek-R1 | 0.97 | 1.00 | -0.03 |
299
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
300
+ | GPT-4o-mini | 0.98 | 0.97 | +0.02 |
301
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
302
+ | Gemini-2.5-flash | 0.20 | 0.32 | -0.12 |
303
+ | Gemini-2.5-flash-nothinking | 0.10 | 0.17 | -0.07 |
304
+ | Qwen3-235b | 1.00 | 0.97 | +0.03 |
305
 
306
  ## any_order_match
307
 
308
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
309
  | --- | --- | --- | --- |
310
+ | DeepSeek-R1 | 1.00 | 1.00 | +0.00 |
311
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
312
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
313
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
314
+ | Gemini-2.5-flash | 0.25 | 0.40 | -0.15 |
315
+ | Gemini-2.5-flash-nothinking | 0.10 | 0.17 | -0.07 |
316
+ | Qwen3-235b | 1.00 | 0.97 | +0.03 |
317
 
318
  ## precision
319
 
320
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
321
  | --- | --- | --- | --- |
322
+ | DeepSeek-R1 | 0.99 | 1.00 | -0.01 |
323
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
324
+ | GPT-4o-mini | 1.00 | 0.99 | +0.00 |
325
  | GPT-5 | 1.00 | 1.00 | +0.00 |
326
+ | Gemini-2.5-flash | 0.88 | 0.86 | +0.02 |
327
+ | Gemini-2.5-flash-nothinking | 0.95 | 0.97 | -0.02 |
328
+ | Qwen3-235b | 1.00 | 1.00 | +0.00 |
329
 
330
  ## recall
331
 
332
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
333
  | --- | --- | --- | --- |
334
+ | DeepSeek-R1 | 1.00 | 1.00 | +0.00 |
335
+ | DeepSeek-V3-1 | 1.00 | 1.00 | +0.00 |
336
+ | GPT-4o-mini | 1.00 | 1.00 | +0.00 |
337
+ | GPT-5 | 1.00 | 1.00 | +0.00 |
338
+ | Gemini-2.5-flash | 0.86 | 0.89 | -0.03 |
339
+ | Gemini-2.5-flash-nothinking | 0.73 | 0.74 | -0.01 |
340
+ | Qwen3-235b | 1.00 | 0.99 | +0.01 |
341
 
342
  ## retry_rate
343
 
344
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
345
  | --- | --- | --- | --- |
346
+ | DeepSeek-R1 | 0.00 | 0.00 | +0.00 |
347
+ | DeepSeek-V3-1 | 0.00 | 0.00 | +0.00 |
348
  | GPT-4o-mini | 0.00 | 0.00 | +0.00 |
349
  | GPT-5 | 0.00 | 0.00 | +0.00 |
350
+ | Gemini-2.5-flash | 0.20 | 0.22 | -0.02 |
351
+ | Gemini-2.5-flash-nothinking | 0.07 | 0.05 | +0.02 |
352
  | Qwen3-235b | 0.00 | 0.00 | +0.00 |
353
 
354
+ ## average_score
355
 
356
  | Model | MCP | Hardcoded | Diff (MCP-Hardcoded) |
357
  | --- | --- | --- | --- |
358
+ | DeepSeek-R1 | 0.79 | 0.82 | -0.03 |
359
+ | DeepSeek-V3-1 | 0.84 | 0.85 | -0.01 |
360
+ | GPT-4o-mini | 0.84 | 0.81 | +0.03 |
361
+ | GPT-5 | 0.70 | 0.72 | -0.03 |
362
+ | Gemini-2.5-flash | 0.06 | 0.11 | -0.05 |
363
+ | Gemini-2.5-flash-nothinking | 0.18 | 0.10 | +0.08 |
364
+ | Qwen3-235b | 0.83 | 0.78 | +0.04 |
365
 
data/processed/RQ1/RadarCharts/Comparison_Powerful_vs_Lightweight.md CHANGED
@@ -2,15 +2,17 @@
2
 
3
  ## Overall Summary (Averaged Across All Project Series)
4
 
 
 
5
  | Metric | GPT-5 | GPT-4o-mini | Diff (A-B) |
6
  | --- | --- | --- | --- |
7
- | exact_match | 0.19 | 0.20 | -0.01 |
8
- | any_order_match | 0.25 | 0.58 | -0.32 |
9
- | precision | 0.92 | 0.83 | +0.09 |
10
- | recall | 0.89 | 0.95 | -0.06 |
11
- | retry_rate | 0.10 | 0.11 | -0.01 |
12
- | average_score | 0.56 | 0.68 | -0.12 |
13
- | pass_rate | 0.68 | 0.71 | -0.03 |
14
 
15
  ---
16
 
@@ -19,111 +21,81 @@
19
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
20
  | --- | --- | --- | --- |
21
  | BookWriter | 0.00 | 0.00 | +0.00 |
22
- | BookWriter-H_A2A | 0.00 | 0.00 | +0.00 |
23
- | EmailResponder | 0.01 | 0.43 | -0.43 |
24
  | GameBuilder | 1.00 | 0.88 | +0.12 |
25
- | LandingPageGenerator | 0.44 | 0.06 | +0.38 |
26
- | LandingPageGenerator-H_A2A | 0.00 | 0.00 | +0.00 |
27
  | MarkdownValidator | 1.00 | 0.97 | +0.03 |
28
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
29
- | RecruitmentAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
30
  | SQLAssistant | 0.00 | 0.00 | +0.00 |
31
- | SQLAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
32
- | SocialMediaManager | 0.03 | 0.05 | -0.02 |
33
- | SocialMediaManager-H_A2A | 0.00 | 0.27 | -0.27 |
34
 
35
  ## any_order_match
36
 
37
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
38
  | --- | --- | --- | --- |
39
  | BookWriter | 0.00 | 0.00 | +0.00 |
40
- | BookWriter-H_A2A | 0.00 | 0.00 | +0.00 |
41
  | EmailResponder | 0.01 | 0.47 | -0.47 |
42
  | GameBuilder | 1.00 | 1.00 | +0.00 |
43
- | LandingPageGenerator | 0.63 | 0.97 | -0.34 |
44
- | LandingPageGenerator-H_A2A | 0.00 | 1.00 | -1.00 |
45
  | MarkdownValidator | 1.00 | 1.00 | +0.00 |
46
- | RecruitmentAssistant | 0.00 | 0.69 | -0.69 |
47
- | RecruitmentAssistant-H_A2A | 0.00 | 0.72 | -0.72 |
48
- | SQLAssistant | 0.02 | 0.32 | -0.31 |
49
- | SQLAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
50
- | SocialMediaManager | 0.48 | 0.84 | -0.36 |
51
- | SocialMediaManager-H_A2A | 0.15 | 0.45 | -0.30 |
52
 
53
  ## precision
54
 
55
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
56
  | --- | --- | --- | --- |
57
- | BookWriter | 0.98 | 0.87 | +0.11 |
58
- | BookWriter-H_A2A | 0.95 | 0.80 | +0.15 |
59
  | EmailResponder | 1.00 | 0.99 | +0.01 |
60
  | GameBuilder | 1.00 | 0.96 | +0.04 |
61
- | LandingPageGenerator | 0.90 | 0.78 | +0.12 |
62
- | LandingPageGenerator-H_A2A | 0.79 | 0.78 | +0.02 |
63
  | MarkdownValidator | 1.00 | 0.99 | +0.01 |
64
- | RecruitmentAssistant | 0.99 | 0.77 | +0.22 |
65
- | RecruitmentAssistant-H_A2A | 0.97 | 0.75 | +0.22 |
66
- | SQLAssistant | 0.77 | 0.73 | +0.05 |
67
- | SQLAssistant-H_A2A | 0.91 | 0.86 | +0.05 |
68
- | SocialMediaManager | 0.84 | 0.63 | +0.20 |
69
- | SocialMediaManager-H_A2A | 0.92 | 0.90 | +0.02 |
70
 
71
  ## recall
72
 
73
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
74
  | --- | --- | --- | --- |
75
- | BookWriter | 0.81 | 0.86 | -0.04 |
76
- | BookWriter-H_A2A | 0.67 | 0.77 | -0.10 |
77
  | EmailResponder | 0.83 | 0.91 | -0.08 |
78
  | GameBuilder | 1.00 | 1.00 | +0.00 |
79
- | LandingPageGenerator | 0.99 | 1.00 | -0.01 |
80
- | LandingPageGenerator-H_A2A | 0.94 | 1.00 | -0.06 |
81
  | MarkdownValidator | 1.00 | 1.00 | +0.00 |
82
  | RecruitmentAssistant | 0.90 | 0.98 | -0.08 |
83
- | RecruitmentAssistant-H_A2A | 0.91 | 0.99 | -0.08 |
84
- | SQLAssistant | 0.87 | 0.95 | -0.09 |
85
- | SQLAssistant-H_A2A | 0.86 | 0.92 | -0.05 |
86
- | SocialMediaManager | 0.88 | 0.99 | -0.11 |
87
- | SocialMediaManager-H_A2A | 0.87 | 0.95 | -0.08 |
88
 
89
  ## retry_rate
90
 
91
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
92
  | --- | --- | --- | --- |
93
  | BookWriter | 0.00 | 0.00 | +0.00 |
94
- | BookWriter-H_A2A | 0.00 | 0.00 | +0.00 |
95
  | EmailResponder | 0.00 | 0.00 | +0.00 |
96
  | GameBuilder | 0.00 | 0.00 | +0.00 |
97
  | LandingPageGenerator | 0.00 | 0.00 | +0.00 |
98
- | LandingPageGenerator-H_A2A | 0.00 | 0.00 | +0.00 |
99
  | MarkdownValidator | 0.00 | 0.00 | +0.00 |
100
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
101
- | RecruitmentAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
102
- | SQLAssistant | 0.54 | 0.41 | +0.13 |
103
- | SQLAssistant-H_A2A | 0.23 | 0.28 | -0.05 |
104
- | SocialMediaManager | 0.47 | 0.63 | -0.17 |
105
- | SocialMediaManager-H_A2A | 0.08 | 0.15 | -0.07 |
106
 
107
  ## average_score
108
 
109
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
110
  | --- | --- | --- | --- |
111
- | LandingPageGenerator | 0.55 | 0.72 | -0.17 |
112
  | MarkdownValidator | 0.71 | 0.82 | -0.11 |
113
- | SQLAssistant | 0.41 | 0.49 | -0.08 |
114
 
115
  ## pass_rate
116
 
117
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
118
  | --- | --- | --- | --- |
119
  | BookWriter | 1.00 | 1.00 | +0.00 |
120
- | BookWriter-H_A2A | 1.00 | 1.00 | +0.00 |
121
- | EmailResponder | 0.00 | 0.00 | +0.00 |
122
- | GameBuilder | 0.00 | 0.00 | +0.00 |
123
- | LandingPageGenerator-H_A2A | 0.32 | 0.59 | -0.27 |
124
  | RecruitmentAssistant | 1.00 | 1.00 | +0.00 |
125
- | RecruitmentAssistant-H_A2A | 1.00 | 1.00 | +0.00 |
126
- | SQLAssistant-H_A2A | 0.53 | 0.58 | -0.05 |
127
- | SocialMediaManager | 0.97 | 0.94 | +0.03 |
128
- | SocialMediaManager-H_A2A | 1.00 | 1.00 | +0.00 |
129
 
 
2
 
3
  ## Overall Summary (Averaged Across All Project Series)
4
 
5
+ _Note: average_score is averaged only across GT projects; pass_rate is averaged only across non-GT projects._
6
+
7
  | Metric | GPT-5 | GPT-4o-mini | Diff (A-B) |
8
  | --- | --- | --- | --- |
9
+ | exact_match | 0.29 | 0.31 | -0.02 |
10
+ | any_order_match | 0.35 | 0.64 | -0.28 |
11
+ | precision | 0.94 | 0.85 | +0.09 |
12
+ | recall | 0.90 | 0.95 | -0.05 |
13
+ | retry_rate | 0.10 | 0.10 | -0.01 |
14
+ | average_score | 0.55 | 0.67 | -0.13 |
15
+ | pass_rate | 1.00 | 0.99 | +0.00 |
16
 
17
  ---
18
 
 
21
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
22
  | --- | --- | --- | --- |
23
  | BookWriter | 0.00 | 0.00 | +0.00 |
24
+ | EmailResponder | 0.01 | 0.43 | -0.42 |
 
25
  | GameBuilder | 1.00 | 0.88 | +0.12 |
26
+ | LandingPageGenerator | 0.29 | 0.04 | +0.26 |
 
27
  | MarkdownValidator | 1.00 | 0.97 | +0.03 |
28
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
 
29
  | SQLAssistant | 0.00 | 0.00 | +0.00 |
30
+ | SocialMediaManager | 0.02 | 0.12 | -0.11 |
 
 
31
 
32
  ## any_order_match
33
 
34
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
35
  | --- | --- | --- | --- |
36
  | BookWriter | 0.00 | 0.00 | +0.00 |
 
37
  | EmailResponder | 0.01 | 0.47 | -0.47 |
38
  | GameBuilder | 1.00 | 1.00 | +0.00 |
39
+ | LandingPageGenerator | 0.42 | 0.98 | -0.56 |
 
40
  | MarkdownValidator | 1.00 | 1.00 | +0.00 |
41
+ | RecruitmentAssistant | 0.00 | 0.70 | -0.70 |
42
+ | SQLAssistant | 0.01 | 0.22 | -0.21 |
43
+ | SocialMediaManager | 0.37 | 0.71 | -0.34 |
 
 
 
44
 
45
  ## precision
46
 
47
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
48
  | --- | --- | --- | --- |
49
+ | BookWriter | 0.97 | 0.85 | +0.12 |
 
50
  | EmailResponder | 1.00 | 0.99 | +0.01 |
51
  | GameBuilder | 1.00 | 0.96 | +0.04 |
52
+ | LandingPageGenerator | 0.86 | 0.78 | +0.08 |
 
53
  | MarkdownValidator | 1.00 | 0.99 | +0.01 |
54
+ | RecruitmentAssistant | 0.98 | 0.76 | +0.22 |
55
+ | SQLAssistant | 0.82 | 0.77 | +0.05 |
56
+ | SocialMediaManager | 0.86 | 0.72 | +0.14 |
 
 
 
57
 
58
  ## recall
59
 
60
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
61
  | --- | --- | --- | --- |
62
+ | BookWriter | 0.77 | 0.83 | -0.06 |
 
63
  | EmailResponder | 0.83 | 0.91 | -0.08 |
64
  | GameBuilder | 1.00 | 1.00 | +0.00 |
65
+ | LandingPageGenerator | 0.97 | 1.00 | -0.03 |
 
66
  | MarkdownValidator | 1.00 | 1.00 | +0.00 |
67
  | RecruitmentAssistant | 0.90 | 0.98 | -0.08 |
68
+ | SQLAssistant | 0.86 | 0.94 | -0.08 |
69
+ | SocialMediaManager | 0.88 | 0.97 | -0.10 |
 
 
 
70
 
71
  ## retry_rate
72
 
73
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
74
  | --- | --- | --- | --- |
75
  | BookWriter | 0.00 | 0.00 | +0.00 |
 
76
  | EmailResponder | 0.00 | 0.00 | +0.00 |
77
  | GameBuilder | 0.00 | 0.00 | +0.00 |
78
  | LandingPageGenerator | 0.00 | 0.00 | +0.00 |
 
79
  | MarkdownValidator | 0.00 | 0.00 | +0.00 |
80
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
81
+ | SQLAssistant | 0.44 | 0.37 | +0.07 |
82
+ | SocialMediaManager | 0.34 | 0.47 | -0.13 |
 
 
 
83
 
84
  ## average_score
85
 
86
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
87
  | --- | --- | --- | --- |
88
+ | LandingPageGenerator | 0.47 | 0.67 | -0.20 |
89
  | MarkdownValidator | 0.71 | 0.82 | -0.11 |
90
+ | SQLAssistant | 0.45 | 0.52 | -0.07 |
91
 
92
  ## pass_rate
93
 
94
  | Project Series | GPT-5 | GPT-4o-mini | Diff (A-B) |
95
  | --- | --- | --- | --- |
96
  | BookWriter | 1.00 | 1.00 | +0.00 |
97
+ | EmailResponder | 1.00 | 1.00 | +0.00 |
98
+ | GameBuilder | 1.00 | 1.00 | +0.00 |
 
 
99
  | RecruitmentAssistant | 1.00 | 1.00 | +0.00 |
100
+ | SocialMediaManager | 0.98 | 0.96 | +0.02 |
 
 
 
101
 
data/processed/RQ1/RadarCharts/Comparison_Reasoning_vs_NonReasoning.md CHANGED
@@ -2,15 +2,17 @@
2
 
3
  ## Overall Summary (Averaged Across All Project Series)
4
 
 
 
5
  | Metric | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
6
  | --- | --- | --- | --- |
7
- | exact_match | 0.15 | 0.22 | -0.07 |
8
- | any_order_match | 0.24 | 0.65 | -0.41 |
9
- | precision | 0.79 | 0.85 | -0.07 |
10
- | recall | 0.90 | 0.96 | -0.06 |
11
- | retry_rate | 0.25 | 0.12 | +0.13 |
12
- | average_score | 0.49 | 0.66 | -0.17 |
13
- | pass_rate | 0.65 | 0.74 | -0.09 |
14
 
15
  ---
16
 
@@ -18,114 +20,84 @@
18
 
19
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
20
  | --- | --- | --- | --- |
21
- | BookWriter | 0.00 | 0.07 | -0.07 |
22
- | BookWriter-H_A2A | 0.00 | 0.00 | +0.00 |
23
  | EmailResponder | 0.00 | 0.16 | -0.16 |
24
  | GameBuilder | 0.94 | 0.90 | +0.04 |
25
- | LandingPageGenerator | 0.00 | 0.36 | -0.36 |
26
- | LandingPageGenerator-H_A2A | 0.00 | 0.00 | +0.00 |
27
  | MarkdownValidator | 0.98 | 1.00 | -0.02 |
28
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
29
- | RecruitmentAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
30
  | SQLAssistant | 0.00 | 0.00 | +0.00 |
31
- | SQLAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
32
- | SocialMediaManager | 0.00 | 0.15 | -0.15 |
33
- | SocialMediaManager-H_A2A | 0.02 | 0.20 | -0.18 |
34
 
35
  ## any_order_match
36
 
37
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
38
  | --- | --- | --- | --- |
39
- | BookWriter | 0.02 | 0.58 | -0.57 |
40
- | BookWriter-H_A2A | 0.00 | 0.58 | -0.58 |
41
  | EmailResponder | 0.03 | 0.24 | -0.21 |
42
  | GameBuilder | 0.96 | 1.00 | -0.04 |
43
- | LandingPageGenerator | 0.18 | 1.00 | -0.82 |
44
- | LandingPageGenerator-H_A2A | 0.03 | 0.98 | -0.95 |
45
  | MarkdownValidator | 1.00 | 1.00 | +0.00 |
46
- | RecruitmentAssistant | 0.04 | 0.03 | +0.01 |
47
- | RecruitmentAssistant-H_A2A | 0.25 | 1.00 | -0.75 |
48
- | SQLAssistant | 0.07 | 0.26 | -0.19 |
49
- | SQLAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
50
- | SocialMediaManager | 0.17 | 0.89 | -0.72 |
51
- | SocialMediaManager-H_A2A | 0.32 | 0.83 | -0.52 |
52
 
53
  ## precision
54
 
55
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
56
  | --- | --- | --- | --- |
57
- | BookWriter | 0.87 | 0.88 | -0.01 |
58
- | BookWriter-H_A2A | 0.91 | 0.92 | -0.01 |
59
  | EmailResponder | 0.77 | 0.98 | -0.21 |
60
  | GameBuilder | 1.00 | 0.97 | +0.03 |
61
- | LandingPageGenerator | 0.78 | 0.90 | -0.12 |
62
- | LandingPageGenerator-H_A2A | 0.78 | 0.77 | +0.01 |
63
  | MarkdownValidator | 1.00 | 1.00 | -0.00 |
64
- | RecruitmentAssistant | 0.81 | 0.85 | -0.04 |
65
- | RecruitmentAssistant-H_A2A | 0.89 | 0.82 | +0.08 |
66
- | SQLAssistant | 0.46 | 0.54 | -0.09 |
67
- | SQLAssistant-H_A2A | 0.64 | 0.89 | -0.25 |
68
- | SocialMediaManager | 0.61 | 0.76 | -0.15 |
69
- | SocialMediaManager-H_A2A | 0.70 | 0.83 | -0.13 |
70
 
71
  ## recall
72
 
73
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
74
  | --- | --- | --- | --- |
75
- | BookWriter | 0.82 | 0.99 | -0.17 |
76
- | BookWriter-H_A2A | 0.77 | 0.95 | -0.18 |
77
  | EmailResponder | 0.89 | 0.88 | +0.02 |
78
  | GameBuilder | 0.99 | 1.00 | -0.01 |
79
  | LandingPageGenerator | 0.94 | 1.00 | -0.06 |
80
- | LandingPageGenerator-H_A2A | 0.93 | 1.00 | -0.07 |
81
  | MarkdownValidator | 1.00 | 1.00 | +0.00 |
82
- | RecruitmentAssistant | 0.88 | 0.95 | -0.07 |
83
- | RecruitmentAssistant-H_A2A | 0.93 | 1.00 | -0.07 |
84
- | SQLAssistant | 0.88 | 0.90 | -0.03 |
85
- | SQLAssistant-H_A2A | 0.89 | 0.90 | -0.01 |
86
- | SocialMediaManager | 0.88 | 0.98 | -0.10 |
87
- | SocialMediaManager-H_A2A | 0.90 | 0.96 | -0.06 |
88
 
89
  ## retry_rate
90
 
91
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
92
  | --- | --- | --- | --- |
93
- | BookWriter | 0.13 | 0.01 | +0.12 |
94
- | BookWriter-H_A2A | 0.02 | 0.00 | +0.02 |
95
  | EmailResponder | 0.16 | 0.02 | +0.14 |
96
  | GameBuilder | 0.00 | 0.00 | +0.00 |
97
- | LandingPageGenerator | 0.09 | 0.00 | +0.09 |
98
- | LandingPageGenerator-H_A2A | 0.00 | 0.03 | -0.03 |
99
  | MarkdownValidator | 0.00 | 0.00 | +0.00 |
100
  | RecruitmentAssistant | 0.05 | 0.00 | +0.05 |
101
- | RecruitmentAssistant-H_A2A | 0.05 | 0.00 | +0.05 |
102
- | SQLAssistant | 0.87 | 0.78 | +0.08 |
103
- | SQLAssistant-H_A2A | 0.67 | 0.27 | +0.40 |
104
- | SocialMediaManager | 0.65 | 0.31 | +0.34 |
105
- | SocialMediaManager-H_A2A | 0.55 | 0.18 | +0.37 |
106
 
107
  ## average_score
108
 
109
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
110
  | --- | --- | --- | --- |
111
- | LandingPageGenerator | 0.33 | 0.85 | -0.52 |
112
  | MarkdownValidator | 0.81 | 0.84 | -0.04 |
113
- | SQLAssistant | 0.34 | 0.30 | +0.04 |
114
 
115
  ## pass_rate
116
 
117
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
118
  | --- | --- | --- | --- |
119
  | BookWriter | 1.00 | 1.00 | +0.00 |
120
- | BookWriter-H_A2A | 1.00 | 1.00 | +0.00 |
121
- | EmailResponder | 0.00 | 0.00 | +0.00 |
122
- | GameBuilder | 0.00 | 0.00 | +0.00 |
123
- | LandingPageGenerator-H_A2A | 0.10 | 0.88 | -0.77 |
124
  | RecruitmentAssistant | 0.99 | 1.00 | -0.01 |
125
- | RecruitmentAssistant-H_A2A | 1.00 | 1.00 | +0.00 |
126
- | SQLAssistant-H_A2A | 0.52 | 0.53 | -0.01 |
127
- | SocialMediaManager | 0.93 | 1.00 | -0.07 |
128
- | SocialMediaManager-H_A2A | 0.97 | 1.00 | -0.03 |
129
 
130
 
131
  ---
@@ -134,15 +106,17 @@
134
 
135
  ## Overall Summary (Averaged Across All Project Series)
136
 
 
 
137
  | Metric | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
138
  | --- | --- | --- | --- |
139
- | exact_match | 0.02 | 0.02 | +0.00 |
140
- | any_order_match | 0.03 | 0.02 | +0.01 |
141
- | precision | 0.92 | 0.94 | -0.01 |
142
- | recall | 0.71 | 0.71 | -0.00 |
143
- | retry_rate | 0.15 | 0.12 | +0.03 |
144
- | average_score | 0.08 | 0.09 | -0.01 |
145
- | pass_rate | 0.65 | 0.66 | -0.00 |
146
 
147
  ---
148
 
@@ -151,111 +125,81 @@
151
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
152
  | --- | --- | --- | --- |
153
  | BookWriter | 0.00 | 0.00 | +0.00 |
154
- | BookWriter-H_A2A | 0.00 | 0.00 | +0.00 |
155
- | EmailResponder | 0.01 | 0.05 | -0.04 |
156
  | GameBuilder | 0.00 | 0.02 | -0.02 |
157
  | LandingPageGenerator | 0.00 | 0.00 | +0.00 |
158
- | LandingPageGenerator-H_A2A | 0.00 | 0.00 | +0.00 |
159
  | MarkdownValidator | 0.26 | 0.13 | +0.12 |
160
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
161
- | RecruitmentAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
162
  | SQLAssistant | 0.00 | 0.00 | +0.00 |
163
- | SQLAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
164
  | SocialMediaManager | 0.00 | 0.00 | +0.00 |
165
- | SocialMediaManager-H_A2A | 0.00 | 0.00 | +0.00 |
166
 
167
  ## any_order_match
168
 
169
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
170
  | --- | --- | --- | --- |
171
  | BookWriter | 0.00 | 0.00 | +0.00 |
172
- | BookWriter-H_A2A | 0.00 | 0.00 | +0.00 |
173
  | EmailResponder | 0.01 | 0.06 | -0.06 |
174
  | GameBuilder | 0.00 | 0.06 | -0.06 |
175
  | LandingPageGenerator | 0.00 | 0.00 | +0.00 |
176
- | LandingPageGenerator-H_A2A | 0.00 | 0.00 | +0.00 |
177
  | MarkdownValidator | 0.33 | 0.13 | +0.19 |
178
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
179
- | RecruitmentAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
180
  | SQLAssistant | 0.00 | 0.00 | +0.00 |
181
- | SQLAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
182
  | SocialMediaManager | 0.00 | 0.00 | +0.00 |
183
- | SocialMediaManager-H_A2A | 0.00 | 0.00 | +0.00 |
184
 
185
  ## precision
186
 
187
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
188
  | --- | --- | --- | --- |
189
- | BookWriter | 1.00 | 1.00 | +0.00 |
190
- | BookWriter-H_A2A | 0.93 | 0.93 | -0.00 |
191
  | EmailResponder | 0.97 | 0.96 | +0.01 |
192
  | GameBuilder | 1.00 | 0.99 | +0.01 |
193
- | LandingPageGenerator | 1.00 | 0.99 | +0.01 |
194
- | LandingPageGenerator-H_A2A | 0.91 | 0.87 | +0.05 |
195
  | MarkdownValidator | 0.87 | 0.96 | -0.09 |
196
- | RecruitmentAssistant | 1.00 | 0.99 | +0.01 |
197
- | RecruitmentAssistant-H_A2A | 0.94 | 0.93 | +0.01 |
198
- | SQLAssistant | 0.73 | 0.80 | -0.07 |
199
- | SQLAssistant-H_A2A | 0.81 | 0.84 | -0.03 |
200
- | SocialMediaManager | 0.97 | 0.98 | -0.00 |
201
- | SocialMediaManager-H_A2A | 0.87 | 0.93 | -0.05 |
202
 
203
  ## recall
204
 
205
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
206
  | --- | --- | --- | --- |
207
- | BookWriter | 0.50 | 0.50 | -0.01 |
208
- | BookWriter-H_A2A | 0.57 | 0.57 | -0.00 |
209
  | EmailResponder | 0.74 | 0.81 | -0.07 |
210
  | GameBuilder | 0.75 | 0.80 | -0.04 |
211
- | LandingPageGenerator | 0.70 | 0.73 | -0.03 |
212
- | LandingPageGenerator-H_A2A | 0.87 | 0.85 | +0.02 |
213
  | MarkdownValidator | 0.88 | 0.73 | +0.15 |
214
- | RecruitmentAssistant | 0.59 | 0.64 | -0.05 |
215
- | RecruitmentAssistant-H_A2A | 0.89 | 0.94 | -0.05 |
216
- | SQLAssistant | 0.74 | 0.74 | -0.00 |
217
- | SQLAssistant-H_A2A | 0.83 | 0.80 | +0.03 |
218
- | SocialMediaManager | 0.39 | 0.39 | +0.00 |
219
- | SocialMediaManager-H_A2A | 0.77 | 0.76 | +0.01 |
220
 
221
  ## retry_rate
222
 
223
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
224
  | --- | --- | --- | --- |
225
  | BookWriter | 0.00 | 0.01 | -0.01 |
226
- | BookWriter-H_A2A | 0.00 | 0.00 | +0.00 |
227
  | EmailResponder | 0.10 | 0.12 | -0.02 |
228
  | GameBuilder | 0.00 | 0.04 | -0.04 |
229
- | LandingPageGenerator | 0.00 | 0.04 | -0.04 |
230
- | LandingPageGenerator-H_A2A | 0.00 | 0.00 | +0.00 |
231
  | MarkdownValidator | 0.21 | 0.06 | +0.15 |
232
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
233
- | RecruitmentAssistant-H_A2A | 0.00 | 0.00 | +0.00 |
234
- | SQLAssistant | 0.60 | 0.53 | +0.07 |
235
- | SQLAssistant-H_A2A | 0.50 | 0.37 | +0.13 |
236
- | SocialMediaManager | 0.11 | 0.10 | +0.01 |
237
- | SocialMediaManager-H_A2A | 0.42 | 0.25 | +0.17 |
238
 
239
  ## average_score
240
 
241
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
242
  | --- | --- | --- | --- |
243
- | LandingPageGenerator | 0.01 | 0.01 | -0.01 |
244
  | MarkdownValidator | 0.08 | 0.14 | -0.06 |
245
- | SQLAssistant | 0.16 | 0.12 | +0.04 |
246
 
247
  ## pass_rate
248
 
249
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
250
  | --- | --- | --- | --- |
251
  | BookWriter | 1.00 | 1.00 | +0.00 |
252
- | BookWriter-H_A2A | 1.00 | 1.00 | +0.00 |
253
- | EmailResponder | 0.00 | 0.00 | +0.00 |
254
- | GameBuilder | 0.00 | 0.00 | +0.00 |
255
- | LandingPageGenerator-H_A2A | 0.00 | 0.00 | +0.00 |
256
  | RecruitmentAssistant | 1.00 | 1.00 | +0.00 |
257
- | RecruitmentAssistant-H_A2A | 1.00 | 1.00 | +0.00 |
258
- | SQLAssistant-H_A2A | 0.55 | 0.57 | -0.02 |
259
  | SocialMediaManager | 1.00 | 1.00 | +0.00 |
260
- | SocialMediaManager-H_A2A | 1.00 | 1.00 | +0.00 |
261
 
 
2
 
3
  ## Overall Summary (Averaged Across All Project Series)
4
 
5
+ _Note: average_score is averaged only across GT projects; pass_rate is averaged only across non-GT projects._
6
+
7
  | Metric | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
8
  | --- | --- | --- | --- |
9
+ | exact_match | 0.24 | 0.31 | -0.07 |
10
+ | any_order_match | 0.31 | 0.65 | -0.34 |
11
+ | precision | 0.80 | 0.87 | -0.07 |
12
+ | recall | 0.91 | 0.96 | -0.05 |
13
+ | retry_rate | 0.22 | 0.11 | +0.11 |
14
+ | average_score | 0.49 | 0.69 | -0.20 |
15
+ | pass_rate | 0.99 | 1.00 | -0.01 |
16
 
17
  ---
18
 
 
20
 
21
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
22
  | --- | --- | --- | --- |
23
+ | BookWriter | 0.00 | 0.04 | -0.04 |
 
24
  | EmailResponder | 0.00 | 0.16 | -0.16 |
25
  | GameBuilder | 0.94 | 0.90 | +0.04 |
26
+ | LandingPageGenerator | 0.00 | 0.24 | -0.24 |
 
27
  | MarkdownValidator | 0.98 | 1.00 | -0.02 |
28
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
 
29
  | SQLAssistant | 0.00 | 0.00 | +0.00 |
30
+ | SocialMediaManager | 0.01 | 0.17 | -0.16 |
 
 
31
 
32
  ## any_order_match
33
 
34
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
35
  | --- | --- | --- | --- |
36
+ | BookWriter | 0.01 | 0.58 | -0.57 |
 
37
  | EmailResponder | 0.03 | 0.24 | -0.21 |
38
  | GameBuilder | 0.96 | 1.00 | -0.04 |
39
+ | LandingPageGenerator | 0.13 | 0.99 | -0.86 |
 
40
  | MarkdownValidator | 1.00 | 1.00 | +0.00 |
41
+ | RecruitmentAssistant | 0.11 | 0.36 | -0.24 |
42
+ | SQLAssistant | 0.04 | 0.17 | -0.13 |
43
+ | SocialMediaManager | 0.22 | 0.87 | -0.66 |
 
 
 
44
 
45
  ## precision
46
 
47
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
48
  | --- | --- | --- | --- |
49
+ | BookWriter | 0.88 | 0.89 | -0.01 |
 
50
  | EmailResponder | 0.77 | 0.98 | -0.21 |
51
  | GameBuilder | 1.00 | 0.97 | +0.03 |
52
+ | LandingPageGenerator | 0.78 | 0.86 | -0.08 |
 
53
  | MarkdownValidator | 1.00 | 1.00 | -0.00 |
54
+ | RecruitmentAssistant | 0.84 | 0.84 | -0.00 |
55
+ | SQLAssistant | 0.52 | 0.66 | -0.14 |
56
+ | SocialMediaManager | 0.64 | 0.78 | -0.14 |
 
 
 
57
 
58
  ## recall
59
 
60
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
61
  | --- | --- | --- | --- |
62
+ | BookWriter | 0.81 | 0.98 | -0.17 |
 
63
  | EmailResponder | 0.89 | 0.88 | +0.02 |
64
  | GameBuilder | 0.99 | 1.00 | -0.01 |
65
  | LandingPageGenerator | 0.94 | 1.00 | -0.06 |
 
66
  | MarkdownValidator | 1.00 | 1.00 | +0.00 |
67
+ | RecruitmentAssistant | 0.90 | 0.97 | -0.07 |
68
+ | SQLAssistant | 0.88 | 0.90 | -0.02 |
69
+ | SocialMediaManager | 0.88 | 0.97 | -0.09 |
 
 
 
70
 
71
  ## retry_rate
72
 
73
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
74
  | --- | --- | --- | --- |
75
+ | BookWriter | 0.09 | 0.01 | +0.09 |
 
76
  | EmailResponder | 0.16 | 0.02 | +0.14 |
77
  | GameBuilder | 0.00 | 0.00 | +0.00 |
78
+ | LandingPageGenerator | 0.06 | 0.01 | +0.05 |
 
79
  | MarkdownValidator | 0.00 | 0.00 | +0.00 |
80
  | RecruitmentAssistant | 0.05 | 0.00 | +0.05 |
81
+ | SQLAssistant | 0.80 | 0.61 | +0.19 |
82
+ | SocialMediaManager | 0.62 | 0.27 | +0.35 |
 
 
 
83
 
84
  ## average_score
85
 
86
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
87
  | --- | --- | --- | --- |
88
+ | LandingPageGenerator | 0.25 | 0.86 | -0.60 |
89
  | MarkdownValidator | 0.81 | 0.84 | -0.04 |
90
+ | SQLAssistant | 0.40 | 0.37 | +0.03 |
91
 
92
  ## pass_rate
93
 
94
  | Project Series | DeepSeek-R1 | DeepSeek-V3-1 | Diff (A-B) |
95
  | --- | --- | --- | --- |
96
  | BookWriter | 1.00 | 1.00 | +0.00 |
97
+ | EmailResponder | 0.99 | 1.00 | -0.01 |
98
+ | GameBuilder | 1.00 | 0.98 | +0.02 |
 
 
99
  | RecruitmentAssistant | 0.99 | 1.00 | -0.01 |
100
+ | SocialMediaManager | 0.94 | 1.00 | -0.06 |
 
 
 
101
 
102
 
103
  ---
 
106
 
107
  ## Overall Summary (Averaged Across All Project Series)
108
 
109
+ _Note: average_score is averaged only across GT projects; pass_rate is averaged only across non-GT projects._
110
+
111
  | Metric | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
112
  | --- | --- | --- | --- |
113
+ | exact_match | 0.03 | 0.02 | +0.01 |
114
+ | any_order_match | 0.04 | 0.03 | +0.01 |
115
+ | precision | 0.93 | 0.95 | -0.01 |
116
+ | recall | 0.70 | 0.71 | -0.00 |
117
+ | retry_rate | 0.14 | 0.11 | +0.03 |
118
+ | average_score | 0.13 | 0.14 | -0.01 |
119
+ | pass_rate | 1.00 | 1.00 | -0.00 |
120
 
121
  ---
122
 
 
125
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
126
  | --- | --- | --- | --- |
127
  | BookWriter | 0.00 | 0.00 | +0.00 |
128
+ | EmailResponder | 0.01 | 0.04 | -0.04 |
 
129
  | GameBuilder | 0.00 | 0.02 | -0.02 |
130
  | LandingPageGenerator | 0.00 | 0.00 | +0.00 |
 
131
  | MarkdownValidator | 0.26 | 0.13 | +0.12 |
132
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
 
133
  | SQLAssistant | 0.00 | 0.00 | +0.00 |
 
134
  | SocialMediaManager | 0.00 | 0.00 | +0.00 |
 
135
 
136
  ## any_order_match
137
 
138
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
139
  | --- | --- | --- | --- |
140
  | BookWriter | 0.00 | 0.00 | +0.00 |
 
141
  | EmailResponder | 0.01 | 0.06 | -0.06 |
142
  | GameBuilder | 0.00 | 0.06 | -0.06 |
143
  | LandingPageGenerator | 0.00 | 0.00 | +0.00 |
 
144
  | MarkdownValidator | 0.33 | 0.13 | +0.19 |
145
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
 
146
  | SQLAssistant | 0.00 | 0.00 | +0.00 |
 
147
  | SocialMediaManager | 0.00 | 0.00 | +0.00 |
 
148
 
149
  ## precision
150
 
151
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
152
  | --- | --- | --- | --- |
153
+ | BookWriter | 0.98 | 0.98 | +0.00 |
 
154
  | EmailResponder | 0.97 | 0.96 | +0.01 |
155
  | GameBuilder | 1.00 | 0.99 | +0.01 |
156
+ | LandingPageGenerator | 0.97 | 0.95 | +0.02 |
 
157
  | MarkdownValidator | 0.87 | 0.96 | -0.09 |
158
+ | RecruitmentAssistant | 0.98 | 0.97 | +0.01 |
159
+ | SQLAssistant | 0.76 | 0.82 | -0.06 |
160
+ | SocialMediaManager | 0.94 | 0.96 | -0.02 |
 
 
 
161
 
162
  ## recall
163
 
164
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
165
  | --- | --- | --- | --- |
166
+ | BookWriter | 0.52 | 0.53 | -0.01 |
 
167
  | EmailResponder | 0.74 | 0.81 | -0.07 |
168
  | GameBuilder | 0.75 | 0.80 | -0.04 |
169
+ | LandingPageGenerator | 0.76 | 0.77 | -0.01 |
 
170
  | MarkdownValidator | 0.88 | 0.73 | +0.15 |
171
+ | RecruitmentAssistant | 0.69 | 0.74 | -0.05 |
172
+ | SQLAssistant | 0.77 | 0.76 | +0.01 |
173
+ | SocialMediaManager | 0.52 | 0.51 | +0.01 |
 
 
 
174
 
175
  ## retry_rate
176
 
177
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
178
  | --- | --- | --- | --- |
179
  | BookWriter | 0.00 | 0.01 | -0.01 |
 
180
  | EmailResponder | 0.10 | 0.12 | -0.02 |
181
  | GameBuilder | 0.00 | 0.04 | -0.04 |
182
+ | LandingPageGenerator | 0.00 | 0.03 | -0.03 |
 
183
  | MarkdownValidator | 0.21 | 0.06 | +0.15 |
184
  | RecruitmentAssistant | 0.00 | 0.00 | +0.00 |
185
+ | SQLAssistant | 0.57 | 0.48 | +0.09 |
186
+ | SocialMediaManager | 0.21 | 0.15 | +0.06 |
 
 
 
187
 
188
  ## average_score
189
 
190
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
191
  | --- | --- | --- | --- |
192
+ | LandingPageGenerator | 0.00 | 0.01 | -0.00 |
193
  | MarkdownValidator | 0.08 | 0.14 | -0.06 |
194
+ | SQLAssistant | 0.29 | 0.27 | +0.02 |
195
 
196
  ## pass_rate
197
 
198
  | Project Series | Gemini-2.5-flash | Gemini-2.5-flash-nothinking | Diff (A-B) |
199
  | --- | --- | --- | --- |
200
  | BookWriter | 1.00 | 1.00 | +0.00 |
201
+ | EmailResponder | 0.99 | 1.00 | -0.01 |
202
+ | GameBuilder | 1.00 | 1.00 | +0.00 |
 
 
203
  | RecruitmentAssistant | 1.00 | 1.00 | +0.00 |
 
 
204
  | SocialMediaManager | 1.00 | 1.00 | +0.00 |
 
205
 
data/processed/RQ1/RadarCharts/EmailResponder-MCP_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:54d373300d71e6253fe1b8406ce65bfe890ba34f2e4451aded082b77890ca09a
3
- size 18382
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:82319b0db8093476b91a21f85c27bed9819c7f7c12f806a6a28563d07f45c67a
3
+ size 18722
data/processed/RQ1/RadarCharts/EmailResponder_Series_Average_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:51a8bf69d906d1e2ab9bd501c4963bdc5becebea100cfda1bc29fed5973e6fc8
3
- size 18507
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f560590ef5841252a5eb343e6accc9d341e4e435f57c0332bee8d44c4cbecfe2
3
+ size 18641
data/processed/RQ1/RadarCharts/EmailResponder_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:49e9ad9ae4ec5cac5b46163d0cac8081bc4971e74af0ddbf66fb1c55e17db977
3
- size 18469
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a3d796f43b54ad5e5be1edd2fb1ea3082dff038e57af8391579ca04c4545198
3
+ size 18806
data/processed/RQ1/RadarCharts/GameBuilder-MCP_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b06beed8b0114b670733470116ea3db358369c95170c5df462c80fcc1642358c
3
- size 17899
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:250cb641161793638113d3b4be23a941428cb3815d034c33bec6288df8d27889
3
+ size 18147
data/processed/RQ1/RadarCharts/GameBuilder_Series_Average_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:819850734e505a15eef48dd8f98ac48b0bded688c5ba03386d2d2bbd17775dd9
3
- size 18193
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70a2da74ccf57f910114d86a35342513fec68e6c0c8549ad7d7dfa6926e97c49
3
+ size 18252
data/processed/RQ1/RadarCharts/GameBuilder_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:79d67995b5802b1d4ae17a733a63265a9b54b2e0450f3976f0f0a73539175445
3
- size 18121
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3f06c27070a9b069d10231f785d1a512030d3351106d0bed6f7576f8ef9606b
3
+ size 18417
data/processed/RQ1/RadarCharts/LandingPageGenerator-A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cded14c20b009b59de83839a071af3ca7a2da52e14ff7ecba7e8c5eb637602a9
3
- size 19474
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e283a4e3b66898455d53a5641cdd8a7f468032e825e7d6677410dba6712e63a
3
+ size 19572
data/processed/RQ1/RadarCharts/LandingPageGenerator-H_A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a313052eb1966fa4719914ab50906368ddb3bbdf074cda1a7211ecdd3c65c2d6
3
- size 19399
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:606eeda4856ee1e6b396aee9f0850eb48b0b5267f2d7af99b77ea800f3df7737
3
+ size 19484
data/processed/RQ1/RadarCharts/LandingPageGenerator-MCP_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8620b8f80a74059960fd37dbd14091b07a35cd9e05f925224df71e8926ae3c0c
3
- size 19590
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c45d54c583452d556450852f197f5b2e4a72f44a7de346cac4696b4384dbd66a
3
+ size 19594
data/processed/RQ1/RadarCharts/LandingPageGenerator_Series_Average_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9a4809a49812fab174e384f7127ebb674632e88af91a1a539037560d87e728de
3
- size 19789
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39e9f915b575c973593c41cf04078e3fa61bb370917238ab079e86c95fcf5019
3
+ size 19739
data/processed/RQ1/RadarCharts/MarkdownValidator-MCP_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eb66c55b36236874cf1f86f4ca99a26dd1142092d5cc9f7ba84b351a39bd8692
3
- size 19351
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85ab29d260711210659302dd5d67d8335dc7e93168fa6282811e7e4a4ec4f987
3
+ size 19187
data/processed/RQ1/RadarCharts/MarkdownValidator_Series_Average_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:28c2d81f35928dfa2dd2fd89f003e5d50360e03249704f07db250865fdf24e11
3
- size 19441
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b6c2486f67c77a03b45609e427f58afc1bb7c63f08de8d47985d40c6c0d250e
3
+ size 19115
data/processed/RQ1/RadarCharts/MarkdownValidator_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:54351eaa5fe68c9a18c899d5f81b584b73f82c920020a5f151ecedc7601036b2
3
- size 19357
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f950d188d24702bf10707a12104307d334ad8c1bfcf7ac9fec4d7d1fa3c2c6e6
3
+ size 19195
data/processed/RQ1/RadarCharts/Model_Legend.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:87995800ac65dd63355b90ca0c2b435f4938cb47a4f64248c171ca64c2b7283e
3
  size 16883
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83288982959a945c33345d775bbd7ec017865674864eb226ced527df37f6f7c8
3
  size 16883
data/processed/RQ1/RadarCharts/Model_Legend_horizontal.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:85eff9bd2173f5f51feb68a7c8c641f40a40cbf341093a50176ee8ae7ae051ea
3
  size 16715
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0833597bb1b7972025c1d714c10b3c8e7ad4e511061d1dc67e5d24d15a8486c
3
  size 16715
data/processed/RQ1/RadarCharts/Overall_Average_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d9a05de9b2aea9728fc9b3c311398a2f1f1032686972e005ca6bbf655cc733da
3
- size 19089
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45566e18abbcba3bdecb82bb9474de2410ccdf5402fa8c30e4c158a488e16283
3
+ size 19086
data/processed/RQ1/RadarCharts/RecruitmentAssistant-A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4d2f7e94106a1e2acab6cbdbfac95f29e8a9eb1c3ce185017a0243217edd7287
3
- size 18188
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df38763b3d0257faea18cae23f51dc8926b446fb14d9d5050866f10f59187f4a
3
+ size 18443
data/processed/RQ1/RadarCharts/RecruitmentAssistant-H_A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4a234d1cca8e73d55dedf0126199c3b59597b90dd541f02cae457634ec8feb32
3
- size 18250
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4cceac688835bc2f7afd6d10bc151023d4bf159b4a28b4f4507d11496bb8bced
3
+ size 18534
data/processed/RQ1/RadarCharts/RecruitmentAssistant-MCP_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:67d2260a99fc6a67dfe3bf7305d246f1b8863e27a3aea65d5b4c2e3094e27e07
3
- size 18286
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c76ccaf0d68501098cf7152cf0d39b8819acbd1dbddf58e662838cf8a413202c
3
+ size 18533
data/processed/RQ1/RadarCharts/RecruitmentAssistant_Series_Average_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:01f7047bdc69d6ee35741573c4bb9b0abeefef4777d1dbf99858a344390a58a8
3
  size 18359
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fa043db14b497da71d4b0b5b8aa8e5fb0019c429b938cfe5a628bcaa4056319
3
  size 18359
data/processed/RQ1/RadarCharts/SQLAssistant-A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dde9c1fb231791991b05329b9e68b62ed93c574921d209472f8d46a4eaba117f
3
- size 19696
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f960fe9b801913e81635335c9f6a3e6dab60a3db4fa07c7cfc126f114298ba2
3
+ size 19699
data/processed/RQ1/RadarCharts/SQLAssistant-H_A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:13028d9a20022c3cbfd116fa7acb377ffc6919aff5b088d8436027517d2278d1
3
  size 19561
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b3c9afbdabbcee68732350554888d7f9cbe20e4269ac8282b2dbdd143540934
3
  size 19561
data/processed/RQ1/RadarCharts/SQLAssistant-MCP_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:26d35e21a146e98e50d5a5b7c84ff5ee5c7e6494238a60215641b74d488f7896
3
- size 19779
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84329640c5fa3fa4ac9d071e223b02fe5bb44d12c423340278cb619afcac9eb6
3
+ size 19774
data/processed/RQ1/RadarCharts/SQLAssistant_Series_Average_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ce8114a22bc99d936c92ebb9f2b296f9705abb9c7b0e1706a2d539450b5de86e
3
- size 19812
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78aed82dbd503751c823ca9f450da06e20f21655891197e4db1b011d943853cb
3
+ size 19655
data/processed/RQ1/RadarCharts/SocialMediaManager-A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3b9cfffcb0634fc47adfe8389af02d687dcc77143251c907ce0670484a401b9e
3
- size 18799
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a6d5ebbbb3b7a8891ca9452a6f0965d045d00bac44f9cc23043beee43787cd68
3
+ size 18896
data/processed/RQ1/RadarCharts/SocialMediaManager-H_A2A_radar.pdf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5f02145d11843714ba35b535a92bda34ef9503ee3ca79436a69c8e616a5ceb00
3
- size 18703
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:164a922c634b36c13aea9a4ed57cf5fe93e2bf021d160231bae766547965fa53
3
+ size 18870