e2_result.json 2.1 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899
  1. {
  2. "timestamp": "2026-03-19T19:32:47.538786",
  3. "category_rs_cross": {
  4. "highly_formal": {
  5. "mean": 0.15584515141964855,
  6. "std": 0,
  7. "values": {
  8. "humaneval": 0.15584515141964855
  9. },
  10. "domains": [
  11. "humaneval"
  12. ]
  13. },
  14. "formal": {
  15. "mean": 6.0420482933500805,
  16. "std": 3.418699412288069,
  17. "values": {
  18. "gsm8k": 2.6233488810620114,
  19. "math": 9.46074770563815
  20. },
  21. "domains": [
  22. "gsm8k",
  23. "math"
  24. ]
  25. },
  26. "semi_formal": {
  27. "mean": 1.2816568422497403,
  28. "std": 0,
  29. "values": {
  30. "flores200": 1.2816568422497403
  31. },
  32. "domains": [
  33. "flores200"
  34. ]
  35. },
  36. "open": {
  37. "mean": 11.212123561386544,
  38. "std": 0,
  39. "values": {
  40. "alpaca": 11.212123561386544
  41. },
  42. "domains": [
  43. "alpaca"
  44. ]
  45. }
  46. },
  47. "sorted_by_convergence": [
  48. {
  49. "category": "highly_formal",
  50. "mean_rs_cross": 0.15584515141964855,
  51. "domains": [
  52. "humaneval"
  53. ]
  54. },
  55. {
  56. "category": "semi_formal",
  57. "mean_rs_cross": 1.2816568422497403,
  58. "domains": [
  59. "flores200"
  60. ]
  61. },
  62. {
  63. "category": "formal",
  64. "mean_rs_cross": 6.0420482933500805,
  65. "domains": [
  66. "gsm8k",
  67. "math"
  68. ]
  69. },
  70. {
  71. "category": "open",
  72. "mean_rs_cross": 11.212123561386544,
  73. "domains": [
  74. "alpaca"
  75. ]
  76. }
  77. ],
  78. "formal_vs_open_comparison": {
  79. "formal_mean": 4.07998057937327,
  80. "open_mean": 11.212123561386544,
  81. "difference": 7.132142982013274,
  82. "ratio": 2.7480825810962193
  83. },
  84. "hypothesis_support": {
  85. "supported": true,
  86. "formal_mean": 4.07998057937327,
  87. "open_mean": 11.212123561386544,
  88. "ratio_open_to_formal": 2.7480825810962193,
  89. "effect_strength": "moderate",
  90. "interpretation": "\u5f62\u5f0f\u5316\u9886\u57df\u8c31\u6536\u655b\u663e\u8457\u4f18\u4e8e\u5f00\u653e\u9886\u57df"
  91. },
  92. "raw_rs_cross": {
  93. "gsm8k": 2.6233488810620114,
  94. "math": 9.46074770563815,
  95. "humaneval": 0.15584515141964855,
  96. "alpaca": 11.212123561386544,
  97. "flores200": 1.2816568422497403
  98. }
  99. }