Files changed (32) hide show
  1. results/zero-shot/CerebrumTech__cere-llama-3-8b-tr.json +6 -6
  2. results/zero-shot/Llama-3.3-70B-Instruct.json +18 -18
  3. results/zero-shot/Ministral-8B-Instruct.json +18 -18
  4. results/zero-shot/Mistral-7B-Instruct-v0.3.json +18 -18
  5. results/zero-shot/Mistral-7B-v0.3.json +18 -18
  6. results/zero-shot/Mixtral-8x7B-Instruct-v0.1.json +18 -18
  7. results/zero-shot/Qwen2.5-0.5B-Instruct.json +6 -6
  8. results/zero-shot/Qwen2.5-0.5B.json +18 -18
  9. results/zero-shot/Qwen2.5-1.5B-Instruct.json +6 -6
  10. results/zero-shot/Qwen2.5-1.5B.json +6 -6
  11. results/zero-shot/Qwen2.5-14B-Instruct.json +6 -6
  12. results/zero-shot/Qwen2.5-14B.json +6 -6
  13. results/zero-shot/Qwen2.5-3B-Instruct.json +6 -6
  14. results/zero-shot/Qwen2.5-3B.json +6 -6
  15. results/zero-shot/Qwen2.5-7B-Instruct.json +6 -6
  16. results/zero-shot/Qwen2.5-7B.json +6 -6
  17. results/zero-shot/aya-23-35B.json +6 -6
  18. results/zero-shot/aya-23-8b.json +18 -18
  19. results/zero-shot/aya-expanse-32b.json +6 -6
  20. results/zero-shot/aya-expanse-8b.json +18 -18
  21. results/zero-shot/commencis-7b.json +6 -6
  22. results/zero-shot/kanarya-2b.json +6 -6
  23. results/zero-shot/llama-3-8b-instruct.json +18 -18
  24. results/zero-shot/llama-3-8b.json +18 -18
  25. results/zero-shot/llama-3.1-8b-instruct.json +18 -18
  26. results/zero-shot/llama-3.1-8b.json +18 -18
  27. results/zero-shot/llama-3.2-1b.json +18 -18
  28. results/zero-shot/llama-3.2-3b-instruct.json +18 -18
  29. results/zero-shot/llama-3.2-3b.json +18 -18
  30. results/zero-shot/mistral-7b.json +6 -6
  31. results/zero-shot/trendyol-7b.json +6 -6
  32. results/zero-shot/turna.json +6 -6
results/zero-shot/CerebrumTech__cere-llama-3-8b-tr.json CHANGED
@@ -124,20 +124,20 @@
124
  {
125
  "name": "mnli_tr",
126
  "task": "natural_language_inference",
127
- "acc": 0.3208,
128
- "acc_norm": 0.3151
129
  },
130
  {
131
  "name": "snli_tr",
132
  "task": "natural_language_inference",
133
- "acc": 0.3238,
134
- "acc_norm": 0.3203
135
  },
136
  {
137
  "name": "xnli_tr",
138
  "task": "natural_language_inference",
139
- "acc": 0.3339321357285429,
140
- "acc_norm": 0.32934131736526945
141
  },
142
  {
143
  "name": "news_cat",
 
124
  {
125
  "name": "mnli_tr",
126
  "task": "natural_language_inference",
127
+ "acc": 0.349,
128
+ "acc_norm": 0.355
129
  },
130
  {
131
  "name": "snli_tr",
132
  "task": "natural_language_inference",
133
+ "acc": 0.3368,
134
+ "acc_norm": 0.3404
135
  },
136
  {
137
  "name": "xnli_tr",
138
  "task": "natural_language_inference",
139
+ "acc": 0.3331337325349301,
140
+ "acc_norm": 0.3377245508982036
141
  },
142
  {
143
  "name": "news_cat",
results/zero-shot/Llama-3.3-70B-Instruct.json CHANGED
@@ -84,24 +84,6 @@
84
  "acc": 0.8313492063492064,
85
  "acc_norm": 0.8214285714285714
86
  },
87
- {
88
- "name": "mnli_tr",
89
- "task": "natural_language_inference",
90
- "acc": 0.348,
91
- "acc_norm": 0.3479
92
- },
93
- {
94
- "name": "snli_tr",
95
- "task": "natural_language_inference",
96
- "acc": 0.3381,
97
- "acc_norm": 0.337
98
- },
99
- {
100
- "name": "xnli_tr",
101
- "task": "natural_language_inference",
102
- "acc": 0.46947791164658637,
103
- "acc_norm": 0.46947791164658637
104
- },
105
  {
106
  "name": "news_cat",
107
  "task": "text_classification",
@@ -194,6 +176,24 @@
194
  "task": "multiple_choice",
195
  "acc": 0.6714285714285714,
196
  "acc_norm": 0.6714285714285714
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
197
  }
198
  ]
199
  }
 
84
  "acc": 0.8313492063492064,
85
  "acc_norm": 0.8214285714285714
86
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
87
  {
88
  "name": "news_cat",
89
  "task": "text_classification",
 
176
  "task": "multiple_choice",
177
  "acc": 0.6714285714285714,
178
  "acc_norm": 0.6714285714285714
179
+ },
180
+ {
181
+ "name": "mnli_tr",
182
+ "task": "natural_language_inference",
183
+ "acc": 0.3214,
184
+ "acc_norm": 0.3213
185
+ },
186
+ {
187
+ "name": "snli_tr",
188
+ "task": "natural_language_inference",
189
+ "acc": 0.3173,
190
+ "acc_norm": 0.3218
191
+ },
192
+ {
193
+ "name": "xnli_tr",
194
+ "task": "natural_language_inference",
195
+ "acc": 0.3333333333333333,
196
+ "acc_norm": 0.3333333333333333
197
  }
198
  ]
199
  }
results/zero-shot/Ministral-8B-Instruct.json CHANGED
@@ -81,24 +81,6 @@
81
  "acc": 0.20294784580498867,
82
  "acc_norm": 0.2032312925170068
83
  },
84
- {
85
- "name": "mnli_tr",
86
- "task": "natural_language_inference",
87
- "acc": 0.3215,
88
- "acc_norm": 0.3297
89
- },
90
- {
91
- "name": "snli_tr",
92
- "task": "natural_language_inference",
93
- "acc": 0.3233,
94
- "acc_norm": 0.247
95
- },
96
- {
97
- "name": "xnli_tr",
98
- "task": "natural_language_inference",
99
- "acc": 0.43815261044176707,
100
- "acc_norm": 0.43815261044176707
101
- },
102
  {
103
  "name": "news_cat",
104
  "task": "text_classification",
@@ -191,6 +173,24 @@
191
  "task": "multiple_choice",
192
  "acc": 0.5857142857142857,
193
  "acc_norm": 0.5857142857142857
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
  ]
196
  }
 
81
  "acc": 0.20294784580498867,
82
  "acc_norm": 0.2032312925170068
83
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
84
  {
85
  "name": "news_cat",
86
  "task": "text_classification",
 
173
  "task": "multiple_choice",
174
  "acc": 0.5857142857142857,
175
  "acc_norm": 0.5857142857142857
176
+ },
177
+ {
178
+ "name": "mnli_tr",
179
+ "task": "natural_language_inference",
180
+ "acc": 0.348,
181
+ "acc_norm": 0.3404
182
+ },
183
+ {
184
+ "name": "snli_tr",
185
+ "task": "natural_language_inference",
186
+ "acc": 0.3371,
187
+ "acc_norm": 0.4137
188
+ },
189
+ {
190
+ "name": "xnli_tr",
191
+ "task": "natural_language_inference",
192
+ "acc": 0.3339321357285429,
193
+ "acc_norm": 0.32275449101796405
194
  }
195
  ]
196
  }
results/zero-shot/Mistral-7B-Instruct-v0.3.json CHANGED
@@ -81,24 +81,6 @@
81
  "acc": 0.45209750566893425,
82
  "acc_norm": 0.7831632653061225
83
  },
84
- {
85
- "name": "mnli_tr",
86
- "task": "natural_language_inference",
87
- "acc": 0.3,
88
- "acc_norm": 0.3128
89
- },
90
- {
91
- "name": "snli_tr",
92
- "task": "natural_language_inference",
93
- "acc": 0.3227,
94
- "acc_norm": 0.323
95
- },
96
- {
97
- "name": "xnli_tr",
98
- "task": "natural_language_inference",
99
- "acc": 0.42650602409638555,
100
- "acc_norm": 0.42650602409638555
101
- },
102
  {
103
  "name": "news_cat",
104
  "task": "text_classification",
@@ -191,6 +173,24 @@
191
  "task": "multiple_choice",
192
  "acc": 0.5,
193
  "acc_norm": 0.5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
  ]
196
  }
 
81
  "acc": 0.45209750566893425,
82
  "acc_norm": 0.7831632653061225
83
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
84
  {
85
  "name": "news_cat",
86
  "task": "text_classification",
 
173
  "task": "multiple_choice",
174
  "acc": 0.5,
175
  "acc_norm": 0.5
176
+ },
177
+ {
178
+ "name": "mnli_tr",
179
+ "task": "natural_language_inference",
180
+ "acc": 0.3128,
181
+ "acc_norm": 0.3165
182
+ },
183
+ {
184
+ "name": "snli_tr",
185
+ "task": "natural_language_inference",
186
+ "acc": 0.318,
187
+ "acc_norm": 0.3219
188
+ },
189
+ {
190
+ "name": "xnli_tr",
191
+ "task": "natural_language_inference",
192
+ "acc": 0.3203592814371258,
193
+ "acc_norm": 0.3341317365269461
194
  }
195
  ]
196
  }
results/zero-shot/Mistral-7B-v0.3.json CHANGED
@@ -120,24 +120,6 @@
120
  "acc": 0.20691609977324263,
121
  "acc_norm": 0.45691609977324266
122
  },
123
- {
124
- "name": "mnli_tr",
125
- "task": "natural_language_inference",
126
- "acc": 0.3233,
127
- "acc_norm": 0.3227
128
- },
129
- {
130
- "name": "snli_tr",
131
- "task": "natural_language_inference",
132
- "acc": 0.3208,
133
- "acc_norm": 0.317
134
- },
135
- {
136
- "name": "xnli_tr",
137
- "task": "natural_language_inference",
138
- "acc": 0.41365461847389556,
139
- "acc_norm": 0.41365461847389556
140
- },
141
  {
142
  "name": "news_cat",
143
  "task": "text_classification",
@@ -191,6 +173,24 @@
191
  "task": "multiple_choice",
192
  "acc": 0.5714285714285714,
193
  "acc_norm": 0.5714285714285714
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
  ]
196
  }
 
120
  "acc": 0.20691609977324263,
121
  "acc_norm": 0.45691609977324266
122
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
123
  {
124
  "name": "news_cat",
125
  "task": "text_classification",
 
173
  "task": "multiple_choice",
174
  "acc": 0.5714285714285714,
175
  "acc_norm": 0.5714285714285714
176
+ },
177
+ {
178
+ "name": "mnli_tr",
179
+ "task": "natural_language_inference",
180
+ "acc": 0.349,
181
+ "acc_norm": 0.3068
182
+ },
183
+ {
184
+ "name": "snli_tr",
185
+ "task": "natural_language_inference",
186
+ "acc": 0.3344,
187
+ "acc_norm": 0.3253
188
+ },
189
+ {
190
+ "name": "xnli_tr",
191
+ "task": "natural_language_inference",
192
+ "acc": 0.33572854291417165,
193
+ "acc_norm": 0.3207584830339321
194
  }
195
  ]
196
  }
results/zero-shot/Mixtral-8x7B-Instruct-v0.1.json CHANGED
@@ -83,24 +83,6 @@
83
  "acc": 0.6292517006802721,
84
  "acc_norm": 0.7859977324263039
85
  },
86
- {
87
- "name": "mnli_tr",
88
- "task": "natural_language_inference",
89
- "acc": 0.2757,
90
- "acc_norm": 0.3115
91
- },
92
- {
93
- "name": "snli_tr",
94
- "task": "natural_language_inference",
95
- "acc": 0.3078,
96
- "acc_norm": 0.3217
97
- },
98
- {
99
- "name": "xnli_tr",
100
- "task": "natural_language_inference",
101
- "acc": 0.4506024096385542,
102
- "acc_norm": 0.4506024096385542
103
- },
104
  {
105
  "name": "news_cat",
106
  "task": "text_classification",
@@ -193,6 +175,24 @@
193
  "task": "multiple_choice",
194
  "acc": 0.5714285714285714,
195
  "acc_norm": 0.5714285714285714
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
196
  }
197
  ]
198
  }
 
83
  "acc": 0.6292517006802721,
84
  "acc_norm": 0.7859977324263039
85
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
86
  {
87
  "name": "news_cat",
88
  "task": "text_classification",
 
175
  "task": "multiple_choice",
176
  "acc": 0.5714285714285714,
177
  "acc_norm": 0.5714285714285714
178
+ },
179
+ {
180
+ "name": "mnli_tr",
181
+ "task": "natural_language_inference",
182
+ "acc": 0.3479,
183
+ "acc_norm": 0.316
184
+ },
185
+ {
186
+ "name": "snli_tr",
187
+ "task": "natural_language_inference",
188
+ "acc": 0.3216,
189
+ "acc_norm": 0.3214
190
+ },
191
+ {
192
+ "name": "xnli_tr",
193
+ "task": "natural_language_inference",
194
+ "acc": 0.3409181636726547,
195
+ "acc_norm": 0.330938123752495
196
  }
197
  ]
198
  }
results/zero-shot/Qwen2.5-0.5B-Instruct.json CHANGED
@@ -123,20 +123,20 @@
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.3211,
127
- "acc_norm": 0.3212
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
- "acc": 0.3239,
133
- "acc_norm": 0.3237
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
- "acc": 0.36626506024096384,
139
- "acc_norm": 0.36626506024096384
140
  },
141
  {
142
  "name": "news_cat",
 
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.3478,
127
+ "acc_norm": 0.3486
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
+ "acc": 0.337,
133
+ "acc_norm": 0.3371
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
+ "acc": 0.3333333333333333,
139
+ "acc_norm": 0.3333333333333333
140
  },
141
  {
142
  "name": "news_cat",
results/zero-shot/Qwen2.5-0.5B.json CHANGED
@@ -120,24 +120,6 @@
120
  "acc": 0.3373015873015873,
121
  "acc_norm": 0.7845804988662132
122
  },
123
- {
124
- "name": "mnli_tr",
125
- "task": "natural_language_inference",
126
- "acc": 0.3208,
127
- "acc_norm": 0.3211
128
- },
129
- {
130
- "name": "snli_tr",
131
- "task": "natural_language_inference",
132
- "acc": 0.3244,
133
- "acc_norm": 0.3237
134
- },
135
- {
136
- "name": "xnli_tr",
137
- "task": "natural_language_inference",
138
- "acc": 0.3614457831325301,
139
- "acc_norm": 0.3614457831325301
140
- },
141
  {
142
  "name": "news_cat",
143
  "task": "text_classification",
@@ -191,6 +173,24 @@
191
  "task": "multiple_choice",
192
  "acc": 0.4714285714285714,
193
  "acc_norm": 0.4714285714285714
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
  ]
196
  }
 
120
  "acc": 0.3373015873015873,
121
  "acc_norm": 0.7845804988662132
122
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
123
  {
124
  "name": "news_cat",
125
  "task": "text_classification",
 
173
  "task": "multiple_choice",
174
  "acc": 0.4714285714285714,
175
  "acc_norm": 0.4714285714285714
176
+ },
177
+ {
178
+ "name": "mnli_tr",
179
+ "task": "natural_language_inference",
180
+ "acc": 0.3468,
181
+ "acc_norm": 0.346
182
+ },
183
+ {
184
+ "name": "snli_tr",
185
+ "task": "natural_language_inference",
186
+ "acc": 0.3371,
187
+ "acc_norm": 0.336
188
+ },
189
+ {
190
+ "name": "xnli_tr",
191
+ "task": "natural_language_inference",
192
+ "acc": 0.3337325349301397,
193
+ "acc_norm": 0.3345309381237525
194
  }
195
  ]
196
  }
results/zero-shot/Qwen2.5-1.5B-Instruct.json CHANGED
@@ -123,20 +123,20 @@
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.3213,
127
- "acc_norm": 0.3213
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
- "acc": 0.3237,
133
- "acc_norm": 0.324
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
- "acc": 0.38835341365461845,
139
- "acc_norm": 0.38835341365461845
140
  },
141
  {
142
  "name": "news_cat",
 
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.3479,
127
+ "acc_norm": 0.3478
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
+ "acc": 0.3368,
133
+ "acc_norm": 0.3366
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
+ "acc": 0.3333333333333333,
139
+ "acc_norm": 0.3333333333333333
140
  },
141
  {
142
  "name": "news_cat",
results/zero-shot/Qwen2.5-1.5B.json CHANGED
@@ -123,20 +123,20 @@
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.3212,
127
- "acc_norm": 0.3208
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
- "acc": 0.3237,
133
- "acc_norm": 0.3238
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
- "acc": 0.4108433734939759,
139
- "acc_norm": 0.4108433734939759
140
  },
141
  {
142
  "name": "news_cat",
 
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.3479,
127
+ "acc_norm": 0.3481
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
+ "acc": 0.3367,
133
+ "acc_norm": 0.3364
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
+ "acc": 0.3333333333333333,
139
+ "acc_norm": 0.3333333333333333
140
  },
141
  {
142
  "name": "news_cat",
results/zero-shot/Qwen2.5-14B-Instruct.json CHANGED
@@ -86,20 +86,20 @@
86
  {
87
  "name": "mnli_tr",
88
  "task": "natural_language_inference",
89
- "acc": 0.2993,
90
- "acc_norm": 0.4052
91
  },
92
  {
93
  "name": "snli_tr",
94
  "task": "natural_language_inference",
95
- "acc": 0.249,
96
- "acc_norm": 0.4158
97
  },
98
  {
99
  "name": "xnli_tr",
100
  "task": "natural_language_inference",
101
- "acc": 0.4108433734939759,
102
- "acc_norm": 0.4108433734939759
103
  },
104
  {
105
  "name": "news_cat",
 
86
  {
87
  "name": "mnli_tr",
88
  "task": "natural_language_inference",
89
+ "acc": 0.3534,
90
+ "acc_norm": 0.2769
91
  },
92
  {
93
  "name": "snli_tr",
94
  "task": "natural_language_inference",
95
+ "acc": 0.3627,
96
+ "acc_norm": 0.2327
97
  },
98
  {
99
  "name": "xnli_tr",
100
  "task": "natural_language_inference",
101
+ "acc": 0.3411177644710579,
102
+ "acc_norm": 0.2778443113772455
103
  },
104
  {
105
  "name": "news_cat",
results/zero-shot/Qwen2.5-14B.json CHANGED
@@ -86,20 +86,20 @@
86
  {
87
  "name": "mnli_tr",
88
  "task": "natural_language_inference",
89
- "acc": 0.3202,
90
- "acc_norm": 0.3281
91
  },
92
  {
93
  "name": "snli_tr",
94
  "task": "natural_language_inference",
95
- "acc": 0.3227,
96
- "acc_norm": 0.3329
97
  },
98
  {
99
  "name": "xnli_tr",
100
  "task": "natural_language_inference",
101
- "acc": 0.46546184738955826,
102
- "acc_norm": 0.46546184738955826
103
  },
104
  {
105
  "name": "news_cat",
 
86
  {
87
  "name": "mnli_tr",
88
  "task": "natural_language_inference",
89
+ "acc": 0.3482,
90
+ "acc_norm": 0.3389
91
  },
92
  {
93
  "name": "snli_tr",
94
  "task": "natural_language_inference",
95
+ "acc": 0.3373,
96
+ "acc_norm": 0.3271
97
  },
98
  {
99
  "name": "xnli_tr",
100
  "task": "natural_language_inference",
101
+ "acc": 0.3335329341317365,
102
+ "acc_norm": 0.33532934131736525
103
  },
104
  {
105
  "name": "news_cat",
results/zero-shot/Qwen2.5-3B-Instruct.json CHANGED
@@ -123,20 +123,20 @@
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.3202,
127
- "acc_norm": 0.3166
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
- "acc": 0.324,
133
- "acc_norm": 0.3233
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
- "acc": 0.3899598393574297,
139
- "acc_norm": 0.3899598393574297
140
  },
141
  {
142
  "name": "news_cat",
 
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.3484,
127
+ "acc_norm": 0.3503
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
+ "acc": 0.3371,
133
+ "acc_norm": 0.3433
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
+ "acc": 0.3335329341317365,
139
+ "acc_norm": 0.33712574850299404
140
  },
141
  {
142
  "name": "news_cat",
results/zero-shot/Qwen2.5-3B.json CHANGED
@@ -123,20 +123,20 @@
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.3212,
127
- "acc_norm": 0.3212
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
- "acc": 0.3219,
133
- "acc_norm": 0.31
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
- "acc": 0.3887550200803213,
139
- "acc_norm": 0.3887550200803213
140
  },
141
  {
142
  "name": "news_cat",
 
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.348,
127
+ "acc_norm": 0.3539
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
+ "acc": 0.3371,
133
+ "acc_norm": 0.3411
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
+ "acc": 0.3335329341317365,
139
+ "acc_norm": 0.3467065868263473
140
  },
141
  {
142
  "name": "news_cat",
results/zero-shot/Qwen2.5-7B-Instruct.json CHANGED
@@ -123,20 +123,20 @@
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.3128,
127
- "acc_norm": 0.3443
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
- "acc": 0.3019,
133
- "acc_norm": 0.3201
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
- "acc": 0.41044176706827307,
139
- "acc_norm": 0.41044176706827307
140
  },
141
  {
142
  "name": "news_cat",
 
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.3564,
127
+ "acc_norm": 0.3258
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
+ "acc": 0.355,
133
+ "acc_norm": 0.3418
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
+ "acc": 0.3379241516966068,
139
+ "acc_norm": 0.3377245508982036
140
  },
141
  {
142
  "name": "news_cat",
results/zero-shot/Qwen2.5-7B.json CHANGED
@@ -123,20 +123,20 @@
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.3204,
127
- "acc_norm": 0.3466
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
- "acc": 0.3236,
133
- "acc_norm": 0.3272
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
- "acc": 0.41847389558232934,
139
- "acc_norm": 0.41847389558232934
140
  },
141
  {
142
  "name": "news_cat",
 
123
  {
124
  "name": "mnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.3482,
127
+ "acc_norm": 0.3233
128
  },
129
  {
130
  "name": "snli_tr",
131
  "task": "natural_language_inference",
132
+ "acc": 0.3364,
133
+ "acc_norm": 0.332
134
  },
135
  {
136
  "name": "xnli_tr",
137
  "task": "natural_language_inference",
138
+ "acc": 0.3339321357285429,
139
+ "acc_norm": 0.33033932135728544
140
  },
141
  {
142
  "name": "news_cat",
results/zero-shot/aya-23-35B.json CHANGED
@@ -86,20 +86,20 @@
86
  {
87
  "name": "mnli_tr",
88
  "task": "natural_language_inference",
89
- "acc": 0.2957,
90
- "acc_norm": 0.3475
91
  },
92
  {
93
  "name": "snli_tr",
94
  "task": "natural_language_inference",
95
- "acc": 0.2881,
96
- "acc_norm": 0.3364
97
  },
98
  {
99
  "name": "xnli_tr",
100
  "task": "natural_language_inference",
101
- "acc": 0.5060240963855421,
102
- "acc_norm": 0.5060240963855421
103
  },
104
  {
105
  "name": "news_cat",
 
86
  {
87
  "name": "mnli_tr",
88
  "task": "natural_language_inference",
89
+ "acc": 0.3733,
90
+ "acc_norm": 0.3217
91
  },
92
  {
93
  "name": "snli_tr",
94
  "task": "natural_language_inference",
95
+ "acc": 0.3727,
96
+ "acc_norm": 0.3241
97
  },
98
  {
99
  "name": "xnli_tr",
100
  "task": "natural_language_inference",
101
+ "acc": 0.38143712574850297,
102
+ "acc_norm": 0.3337325349301397
103
  },
104
  {
105
  "name": "news_cat",
results/zero-shot/aya-23-8b.json CHANGED
@@ -41,12 +41,6 @@
41
  "exact_match": 0.10017756732761172,
42
  "f1": 0.16569513329103133
43
  },
44
- {
45
- "name": "mnli_tr",
46
- "task": "natural_language_inference",
47
- "acc": 0.3436,
48
- "acc_norm": 0.3477
49
- },
50
  {
51
  "name": "news_cat",
52
  "task": "text_classification",
@@ -65,12 +59,6 @@
65
  "acc": 0.42550274223034734,
66
  "acc_norm": 0.4273308957952468
67
  },
68
- {
69
- "name": "snli_tr",
70
- "task": "natural_language_inference",
71
- "acc": 0.3249,
72
- "acc_norm": 0.3367
73
- },
74
  {
75
  "name": "sts_tr",
76
  "task": "text_classification",
@@ -89,12 +77,6 @@
89
  "acc": 0.596,
90
  "acc_norm": 0.596
91
  },
92
- {
93
- "name": "xnli_tr",
94
- "task": "natural_language_inference",
95
- "acc": 0.4771084337349398,
96
- "acc_norm": 0.4771084337349398
97
- },
98
  {
99
  "name": "xquad_tr",
100
  "task": "extractive_question_answering",
@@ -193,6 +175,24 @@
193
  "task": "multiple_choice",
194
  "acc": 0.6268364348677767,
195
  "acc_norm": 0.6268364348677767
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
196
  }
197
  ]
198
  }
 
41
  "exact_match": 0.10017756732761172,
42
  "f1": 0.16569513329103133
43
  },
 
 
 
 
 
 
44
  {
45
  "name": "news_cat",
46
  "task": "text_classification",
 
59
  "acc": 0.42550274223034734,
60
  "acc_norm": 0.4273308957952468
61
  },
 
 
 
 
 
 
62
  {
63
  "name": "sts_tr",
64
  "task": "text_classification",
 
77
  "acc": 0.596,
78
  "acc_norm": 0.596
79
  },
 
 
 
 
 
 
80
  {
81
  "name": "xquad_tr",
82
  "task": "extractive_question_answering",
 
175
  "task": "multiple_choice",
176
  "acc": 0.6268364348677767,
177
  "acc_norm": 0.6268364348677767
178
+ },
179
+ {
180
+ "name": "mnli_tr",
181
+ "task": "natural_language_inference",
182
+ "acc": 0.3269,
183
+ "acc_norm": 0.3216
184
+ },
185
+ {
186
+ "name": "snli_tr",
187
+ "task": "natural_language_inference",
188
+ "acc": 0.3358,
189
+ "acc_norm": 0.3236
190
+ },
191
+ {
192
+ "name": "xnli_tr",
193
+ "task": "natural_language_inference",
194
+ "acc": 0.33672654690618764,
195
+ "acc_norm": 0.332934131736527
196
  }
197
  ]
198
  }
results/zero-shot/aya-expanse-32b.json CHANGED
@@ -85,20 +85,20 @@
85
  {
86
  "name": "mnli_tr",
87
  "task": "natural_language_inference",
88
- "acc": 0.2444,
89
- "acc_norm": 0.3458
90
  },
91
  {
92
  "name": "snli_tr",
93
  "task": "natural_language_inference",
94
- "acc": 0.1896,
95
- "acc_norm": 0.3355
96
  },
97
  {
98
  "name": "xnli_tr",
99
  "task": "natural_language_inference",
100
- "acc": 0.5056224899598394,
101
- "acc_norm": 0.5056224899598394
102
  },
103
  {
104
  "name": "news_cat",
 
85
  {
86
  "name": "mnli_tr",
87
  "task": "natural_language_inference",
88
+ "acc": 0.4248,
89
+ "acc_norm": 0.3234
90
  },
91
  {
92
  "name": "snli_tr",
93
  "task": "natural_language_inference",
94
+ "acc": 0.4711,
95
+ "acc_norm": 0.325
96
  },
97
  {
98
  "name": "xnli_tr",
99
  "task": "natural_language_inference",
100
+ "acc": 0.40459081836327343,
101
+ "acc_norm": 0.33512974051896205
102
  },
103
  {
104
  "name": "news_cat",
results/zero-shot/aya-expanse-8b.json CHANGED
@@ -51,12 +51,6 @@
51
  "rouge2": 0.21362825588593481,
52
  "rougeL": 0.29773476508614094
53
  },
54
- {
55
- "name": "mnli_tr",
56
- "task": "natural_language_inference",
57
- "acc": 0.3078,
58
- "acc_norm": 0.35
59
- },
60
  {
61
  "name": "news_cat",
62
  "task": "text_classification",
@@ -75,12 +69,6 @@
75
  "acc": 0.5877513711151737,
76
  "acc_norm": 0.579981718464351
77
  },
78
- {
79
- "name": "snli_tr",
80
- "task": "natural_language_inference",
81
- "acc": 0.344,
82
- "acc_norm": 0.3435
83
- },
84
  {
85
  "name": "sts_tr",
86
  "task": "text_classification",
@@ -119,12 +107,6 @@
119
  "rouge2": 0.133428873146516,
120
  "rougeL": 0.2083669711429916
121
  },
122
- {
123
- "name": "xnli_tr",
124
- "task": "natural_language_inference",
125
- "acc": 0.4919678714859438,
126
- "acc_norm": 0.4919678714859438
127
- },
128
  {
129
  "name": "xquad_tr",
130
  "task": "extractive_question_answering",
@@ -191,6 +173,24 @@
191
  "task": "multiple_choice",
192
  "acc": 0.614103819784525,
193
  "acc_norm": 0.614103819784525
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
  ]
196
  }
 
51
  "rouge2": 0.21362825588593481,
52
  "rougeL": 0.29773476508614094
53
  },
 
 
 
 
 
 
54
  {
55
  "name": "news_cat",
56
  "task": "text_classification",
 
69
  "acc": 0.5877513711151737,
70
  "acc_norm": 0.579981718464351
71
  },
 
 
 
 
 
 
72
  {
73
  "name": "sts_tr",
74
  "task": "text_classification",
 
107
  "rouge2": 0.133428873146516,
108
  "rougeL": 0.2083669711429916
109
  },
 
 
 
 
 
 
110
  {
111
  "name": "xquad_tr",
112
  "task": "extractive_question_answering",
 
173
  "task": "multiple_choice",
174
  "acc": 0.614103819784525,
175
  "acc_norm": 0.614103819784525
176
+ },
177
+ {
178
+ "name": "mnli_tr",
179
+ "task": "natural_language_inference",
180
+ "acc": 0.3609,
181
+ "acc_norm": 0.319
182
+ },
183
+ {
184
+ "name": "snli_tr",
185
+ "task": "natural_language_inference",
186
+ "acc": 0.3164,
187
+ "acc_norm": 0.317
188
+ },
189
+ {
190
+ "name": "xnli_tr",
191
+ "task": "natural_language_inference",
192
+ "acc": 0.36347305389221557,
193
+ "acc_norm": 0.3313373253493014
194
  }
195
  ]
196
  }
results/zero-shot/commencis-7b.json CHANGED
@@ -111,20 +111,20 @@
111
  {
112
  "name": "mnli_tr",
113
  "task": "natural_language_inference",
114
- "acc": 0.3058,
115
- "acc_norm": 0.3103
116
  },
117
  {
118
  "name": "snli_tr",
119
  "task": "natural_language_inference",
120
- "acc": 0.2972,
121
- "acc_norm": 0.32
122
  },
123
  {
124
  "name": "xnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.3141716566866267,
127
- "acc_norm": 0.3281437125748503
128
  },
129
  {
130
  "name": "news_cat",
 
111
  {
112
  "name": "mnli_tr",
113
  "task": "natural_language_inference",
114
+ "acc": 0.3329,
115
+ "acc_norm": 0.3179
116
  },
117
  {
118
  "name": "snli_tr",
119
  "task": "natural_language_inference",
120
+ "acc": 0.3148,
121
+ "acc_norm": 0.3207
122
  },
123
  {
124
  "name": "xnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.3269461077844311,
127
+ "acc_norm": 0.33033932135728544
128
  },
129
  {
130
  "name": "news_cat",
results/zero-shot/kanarya-2b.json CHANGED
@@ -53,20 +53,20 @@
53
  {
54
  "name": "mnli_tr",
55
  "task": "natural_language_inference",
56
- "acc": 0.3278,
57
- "acc_norm": 0.3463
58
  },
59
  {
60
  "name": "snli_tr",
61
  "task": "natural_language_inference",
62
- "acc": 0.3088,
63
- "acc_norm": 0.3109
64
  },
65
  {
66
  "name": "xnli_tr",
67
  "task": "natural_language_inference",
68
- "acc": 0.3273453093812375,
69
- "acc_norm": 0.3341317365269461
70
  },
71
  {
72
  "name": "offenseval_tr",
 
53
  {
54
  "name": "mnli_tr",
55
  "task": "natural_language_inference",
56
+ "acc": 0.334,
57
+ "acc_norm": 0.3231
58
  },
59
  {
60
  "name": "snli_tr",
61
  "task": "natural_language_inference",
62
+ "acc": 0.3171,
63
+ "acc_norm": 0.3104
64
  },
65
  {
66
  "name": "xnli_tr",
67
  "task": "natural_language_inference",
68
+ "acc": 0.34231536926147704,
69
+ "acc_norm": 0.33512974051896205
70
  },
71
  {
72
  "name": "offenseval_tr",
results/zero-shot/llama-3-8b-instruct.json CHANGED
@@ -40,12 +40,6 @@
40
  "exact_match": 0.0424681858538029,
41
  "f1": 0.11050423163975964
42
  },
43
- {
44
- "name": "mnli_tr",
45
- "task": "natural_language_inference",
46
- "acc": 0.3201,
47
- "acc_norm": 0.3653
48
- },
49
  {
50
  "name": "news_cat",
51
  "task": "text_classification",
@@ -64,12 +58,6 @@
64
  "acc": 0.603290676416819,
65
  "acc_norm": 0.5790676416819013
66
  },
67
- {
68
- "name": "snli_tr",
69
- "task": "natural_language_inference",
70
- "acc": 0.3283,
71
- "acc_norm": 0.353
72
- },
73
  {
74
  "name": "sts_tr",
75
  "task": "text_classification",
@@ -88,12 +76,6 @@
88
  "acc": 0.586,
89
  "acc_norm": 0.586
90
  },
91
- {
92
- "name": "xnli_tr",
93
- "task": "natural_language_inference",
94
- "acc": 0.4389558232931727,
95
- "acc_norm": 0.4389558232931727
96
- },
97
  {
98
  "name": "xquad_tr",
99
  "task": "extractive_question_answering",
@@ -192,6 +174,24 @@
192
  "task": "multiple_choice",
193
  "acc": 0.6346718903036239,
194
  "acc_norm": 0.6346718903036239
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
195
  }
196
  ]
197
  }
 
40
  "exact_match": 0.0424681858538029,
41
  "f1": 0.11050423163975964
42
  },
 
 
 
 
 
 
43
  {
44
  "name": "news_cat",
45
  "task": "text_classification",
 
58
  "acc": 0.603290676416819,
59
  "acc_norm": 0.5790676416819013
60
  },
 
 
 
 
 
 
61
  {
62
  "name": "sts_tr",
63
  "task": "text_classification",
 
76
  "acc": 0.586,
77
  "acc_norm": 0.586
78
  },
 
 
 
 
 
 
79
  {
80
  "name": "xquad_tr",
81
  "task": "extractive_question_answering",
 
174
  "task": "multiple_choice",
175
  "acc": 0.6346718903036239,
176
  "acc_norm": 0.6346718903036239
177
+ },
178
+ {
179
+ "name": "mnli_tr",
180
+ "task": "natural_language_inference",
181
+ "acc": 0.3507,
182
+ "acc_norm": 0.3029
183
+ },
184
+ {
185
+ "name": "snli_tr",
186
+ "task": "natural_language_inference",
187
+ "acc": 0.3331,
188
+ "acc_norm": 0.3076
189
+ },
190
+ {
191
+ "name": "xnli_tr",
192
+ "task": "natural_language_inference",
193
+ "acc": 0.3311377245508982,
194
+ "acc_norm": 0.2994011976047904
195
  }
196
  ]
197
  }
results/zero-shot/llama-3-8b.json CHANGED
@@ -39,12 +39,6 @@
39
  "exact_match": 0.13465522343888725,
40
  "f1": 0.19144550324599957
41
  },
42
- {
43
- "name": "mnli_tr",
44
- "task": "natural_language_inference",
45
- "acc": 0.3206,
46
- "acc_norm": 0.3329
47
- },
48
  {
49
  "name": "news_cat",
50
  "task": "text_classification",
@@ -63,12 +57,6 @@
63
  "acc": 0.42550274223034734,
64
  "acc_norm": 0.5173674588665448
65
  },
66
- {
67
- "name": "snli_tr",
68
- "task": "natural_language_inference",
69
- "acc": 0.325,
70
- "acc_norm": 0.3766
71
- },
72
  {
73
  "name": "sts_tr",
74
  "task": "text_classification",
@@ -87,12 +75,6 @@
87
  "acc": 0.618,
88
  "acc_norm": 0.618
89
  },
90
- {
91
- "name": "xnli_tr",
92
- "task": "natural_language_inference",
93
- "acc": 0.4839357429718876,
94
- "acc_norm": 0.4839357429718876
95
- },
96
  {
97
  "name": "xquad_tr",
98
  "task": "extractive_question_answering",
@@ -191,6 +173,24 @@
191
  "task": "multiple_choice",
192
  "acc": 0.6278158667972575,
193
  "acc_norm": 0.6278158667972575
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
  ]
196
  }
 
39
  "exact_match": 0.13465522343888725,
40
  "f1": 0.19144550324599957
41
  },
 
 
 
 
 
 
42
  {
43
  "name": "news_cat",
44
  "task": "text_classification",
 
57
  "acc": 0.42550274223034734,
58
  "acc_norm": 0.5173674588665448
59
  },
 
 
 
 
 
 
60
  {
61
  "name": "sts_tr",
62
  "task": "text_classification",
 
75
  "acc": 0.618,
76
  "acc_norm": 0.618
77
  },
 
 
 
 
 
 
78
  {
79
  "name": "xquad_tr",
80
  "task": "extractive_question_answering",
 
173
  "task": "multiple_choice",
174
  "acc": 0.6278158667972575,
175
  "acc_norm": 0.6278158667972575
176
+ },
177
+ {
178
+ "name": "mnli_tr",
179
+ "task": "natural_language_inference",
180
+ "acc": 0.348,
181
+ "acc_norm": 0.3339
182
+ },
183
+ {
184
+ "name": "snli_tr",
185
+ "task": "natural_language_inference",
186
+ "acc": 0.3358,
187
+ "acc_norm": 0.2861
188
+ },
189
+ {
190
+ "name": "xnli_tr",
191
+ "task": "natural_language_inference",
192
+ "acc": 0.3339321357285429,
193
+ "acc_norm": 0.3203592814371258
194
  }
195
  ]
196
  }
results/zero-shot/llama-3.1-8b-instruct.json CHANGED
@@ -39,12 +39,6 @@
39
  "exact_match": 0.09115122817401598,
40
  "f1": 0.15627870028803578
41
  },
42
- {
43
- "name": "mnli_tr",
44
- "task": "natural_language_inference",
45
- "acc": 0.3209,
46
- "acc_norm": 0.3596
47
- },
48
  {
49
  "name": "news_cat",
50
  "task": "text_classification",
@@ -63,12 +57,6 @@
63
  "acc": 0.4648080438756856,
64
  "acc_norm": 0.5648994515539305
65
  },
66
- {
67
- "name": "snli_tr",
68
- "task": "natural_language_inference",
69
- "acc": 0.3028,
70
- "acc_norm": 0.3528
71
- },
72
  {
73
  "name": "sts_tr",
74
  "task": "text_classification",
@@ -87,12 +75,6 @@
87
  "acc": 0.608,
88
  "acc_norm": 0.608
89
  },
90
- {
91
- "name": "xnli_tr",
92
- "task": "natural_language_inference",
93
- "acc": 0.4807228915662651,
94
- "acc_norm": 0.4807228915662651
95
- },
96
  {
97
  "name": "xquad_tr",
98
  "task": "extractive_question_answering",
@@ -191,6 +173,24 @@
191
  "task": "multiple_choice",
192
  "acc": 0.6571988246816847,
193
  "acc_norm": 0.6571988246816847
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
  ]
196
  }
 
39
  "exact_match": 0.09115122817401598,
40
  "f1": 0.15627870028803578
41
  },
 
 
 
 
 
 
42
  {
43
  "name": "news_cat",
44
  "task": "text_classification",
 
57
  "acc": 0.4648080438756856,
58
  "acc_norm": 0.5648994515539305
59
  },
 
 
 
 
 
 
60
  {
61
  "name": "sts_tr",
62
  "task": "text_classification",
 
75
  "acc": 0.608,
76
  "acc_norm": 0.608
77
  },
 
 
 
 
 
 
78
  {
79
  "name": "xquad_tr",
80
  "task": "extractive_question_answering",
 
173
  "task": "multiple_choice",
174
  "acc": 0.6571988246816847,
175
  "acc_norm": 0.6571988246816847
176
+ },
177
+ {
178
+ "name": "mnli_tr",
179
+ "task": "natural_language_inference",
180
+ "acc": 0.3666,
181
+ "acc_norm": 0.3275
182
+ },
183
+ {
184
+ "name": "snli_tr",
185
+ "task": "natural_language_inference",
186
+ "acc": 0.3593,
187
+ "acc_norm": 0.3178
188
+ },
189
+ {
190
+ "name": "xnli_tr",
191
+ "task": "natural_language_inference",
192
+ "acc": 0.3536926147704591,
193
+ "acc_norm": 0.3435129740518962
194
  }
195
  ]
196
  }
results/zero-shot/llama-3.1-8b.json CHANGED
@@ -39,12 +39,6 @@
39
  "exact_match": 0.09248298313110388,
40
  "f1": 0.15127108197296948
41
  },
42
- {
43
- "name": "mnli_tr",
44
- "task": "natural_language_inference",
45
- "acc": 0.3495,
46
- "acc_norm": 0.3481
47
- },
48
  {
49
  "name": "news_cat",
50
  "task": "text_classification",
@@ -63,12 +57,6 @@
63
  "acc": 0.4227605118829982,
64
  "acc_norm": 0.506398537477148
65
  },
66
- {
67
- "name": "snli_tr",
68
- "task": "natural_language_inference",
69
- "acc": 0.3169,
70
- "acc_norm": 0.3379
71
- },
72
  {
73
  "name": "sts_tr",
74
  "task": "text_classification",
@@ -87,12 +75,6 @@
87
  "acc": 0.626,
88
  "acc_norm": 0.626
89
  },
90
- {
91
- "name": "xnli_tr",
92
- "task": "natural_language_inference",
93
- "acc": 0.4947791164658635,
94
- "acc_norm": 0.4947791164658635
95
- },
96
  {
97
  "name": "xquad_tr",
98
  "task": "extractive_question_answering",
@@ -191,6 +173,24 @@
191
  "task": "multiple_choice",
192
  "acc": 0.6317335945151812,
193
  "acc_norm": 0.6317335945151812
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
  ]
196
  }
 
39
  "exact_match": 0.09248298313110388,
40
  "f1": 0.15127108197296948
41
  },
 
 
 
 
 
 
42
  {
43
  "name": "news_cat",
44
  "task": "text_classification",
 
57
  "acc": 0.4227605118829982,
58
  "acc_norm": 0.506398537477148
59
  },
 
 
 
 
 
 
60
  {
61
  "name": "sts_tr",
62
  "task": "text_classification",
 
75
  "acc": 0.626,
76
  "acc_norm": 0.626
77
  },
 
 
 
 
 
 
78
  {
79
  "name": "xquad_tr",
80
  "task": "extractive_question_answering",
 
173
  "task": "multiple_choice",
174
  "acc": 0.6317335945151812,
175
  "acc_norm": 0.6317335945151812
176
+ },
177
+ {
178
+ "name": "mnli_tr",
179
+ "task": "natural_language_inference",
180
+ "acc": 0.3194,
181
+ "acc_norm": 0.3209
182
+ },
183
+ {
184
+ "name": "snli_tr",
185
+ "task": "natural_language_inference",
186
+ "acc": 0.3432,
187
+ "acc_norm": 0.3222
188
+ },
189
+ {
190
+ "name": "xnli_tr",
191
+ "task": "natural_language_inference",
192
+ "acc": 0.3027944111776447,
193
+ "acc_norm": 0.3327345309381238
194
  }
195
  ]
196
  }
results/zero-shot/llama-3.2-1b.json CHANGED
@@ -51,12 +51,6 @@
51
  "rouge2": 0.13426790568610214,
52
  "rougeL": 0.18915548037371513
53
  },
54
- {
55
- "name": "mnli_tr",
56
- "task": "natural_language_inference",
57
- "acc": 0.3232,
58
- "acc_norm": 0.334
59
- },
60
  {
61
  "name": "news_cat",
62
  "task": "text_classification",
@@ -75,12 +69,6 @@
75
  "acc": 0.56672760511883,
76
  "acc_norm": 0.5781535648994516
77
  },
78
- {
79
- "name": "snli_tr",
80
- "task": "natural_language_inference",
81
- "acc": 0.3239,
82
- "acc_norm": 0.3105
83
- },
84
  {
85
  "name": "sts_tr",
86
  "task": "text_classification",
@@ -119,12 +107,6 @@
119
  "rouge2": 0.07190935921365724,
120
  "rougeL": 0.13255123335488528
121
  },
122
- {
123
- "name": "xnli_tr",
124
- "task": "natural_language_inference",
125
- "acc": 0.4389558232931727,
126
- "acc_norm": 0.4389558232931727
127
- },
128
  {
129
  "name": "xquad_tr",
130
  "task": "extractive_question_answering",
@@ -223,6 +205,24 @@
223
  "task": "multiple_choice",
224
  "acc": 0.5768854064642507,
225
  "acc_norm": 0.5768854064642507
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
226
  }
227
  ]
228
  }
 
51
  "rouge2": 0.13426790568610214,
52
  "rougeL": 0.18915548037371513
53
  },
 
 
 
 
 
 
54
  {
55
  "name": "news_cat",
56
  "task": "text_classification",
 
69
  "acc": 0.56672760511883,
70
  "acc_norm": 0.5781535648994516
71
  },
 
 
 
 
 
 
72
  {
73
  "name": "sts_tr",
74
  "task": "text_classification",
 
107
  "rouge2": 0.07190935921365724,
108
  "rougeL": 0.13255123335488528
109
  },
 
 
 
 
 
 
110
  {
111
  "name": "xquad_tr",
112
  "task": "extractive_question_answering",
 
205
  "task": "multiple_choice",
206
  "acc": 0.5768854064642507,
207
  "acc_norm": 0.5768854064642507
208
+ },
209
+ {
210
+ "name": "mnli_tr",
211
+ "task": "natural_language_inference",
212
+ "acc": 0.3504,
213
+ "acc_norm": 0.3542
214
+ },
215
+ {
216
+ "name": "snli_tr",
217
+ "task": "natural_language_inference",
218
+ "acc": 0.3372,
219
+ "acc_norm": 0.3636
220
+ },
221
+ {
222
+ "name": "xnli_tr",
223
+ "task": "natural_language_inference",
224
+ "acc": 0.33532934131736525,
225
+ "acc_norm": 0.3600798403193613
226
  }
227
  ]
228
  }
results/zero-shot/llama-3.2-3b-instruct.json CHANGED
@@ -51,12 +51,6 @@
51
  "rouge2": 0.1480446780314802,
52
  "rougeL": 0.2106440565987865
53
  },
54
- {
55
- "name": "mnli_tr",
56
- "task": "natural_language_inference",
57
- "acc": 0.32,
58
- "acc_norm": 0.3141
59
- },
60
  {
61
  "name": "news_cat",
62
  "task": "text_classification",
@@ -75,12 +69,6 @@
75
  "acc": 0.4227605118829982,
76
  "acc_norm": 0.42413162705667273
77
  },
78
- {
79
- "name": "snli_tr",
80
- "task": "natural_language_inference",
81
- "acc": 0.319,
82
- "acc_norm": 0.2923
83
- },
84
  {
85
  "name": "sts_tr",
86
  "task": "text_classification",
@@ -119,12 +107,6 @@
119
  "rouge2": 0.06736413357191079,
120
  "rougeL": 0.12750762702828333
121
  },
122
- {
123
- "name": "xnli_tr",
124
- "task": "natural_language_inference",
125
- "acc": 0.42811244979919677,
126
- "acc_norm": 0.42811244979919677
127
- },
128
  {
129
  "name": "xquad_tr",
130
  "task": "extractive_question_answering",
@@ -223,6 +205,24 @@
223
  "task": "multiple_choice",
224
  "acc": 0.6366307541625857,
225
  "acc_norm": 0.6366307541625857
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
226
  }
227
  ]
228
  }
 
51
  "rouge2": 0.1480446780314802,
52
  "rougeL": 0.2106440565987865
53
  },
 
 
 
 
 
 
54
  {
55
  "name": "news_cat",
56
  "task": "text_classification",
 
69
  "acc": 0.4227605118829982,
70
  "acc_norm": 0.42413162705667273
71
  },
 
 
 
 
 
 
72
  {
73
  "name": "sts_tr",
74
  "task": "text_classification",
 
107
  "rouge2": 0.06736413357191079,
108
  "rougeL": 0.12750762702828333
109
  },
 
 
 
 
 
 
110
  {
111
  "name": "xquad_tr",
112
  "task": "extractive_question_answering",
 
205
  "task": "multiple_choice",
206
  "acc": 0.6366307541625857,
207
  "acc_norm": 0.6366307541625857
208
+ },
209
+ {
210
+ "name": "mnli_tr",
211
+ "task": "natural_language_inference",
212
+ "acc": 0.3476,
213
+ "acc_norm": 0.3517
214
+ },
215
+ {
216
+ "name": "snli_tr",
217
+ "task": "natural_language_inference",
218
+ "acc": 0.338,
219
+ "acc_norm": 0.3438
220
+ },
221
+ {
222
+ "name": "xnli_tr",
223
+ "task": "natural_language_inference",
224
+ "acc": 0.3341317365269461,
225
+ "acc_norm": 0.33712574850299404
226
  }
227
  ]
228
  }
results/zero-shot/llama-3.2-3b.json CHANGED
@@ -39,12 +39,6 @@
39
  "exact_match": 0.06007694584196508,
40
  "f1": 0.10654344736882515
41
  },
42
- {
43
- "name": "mnli_tr",
44
- "task": "natural_language_inference",
45
- "acc": 0.3488,
46
- "acc_norm": 0.3478
47
- },
48
  {
49
  "name": "news_cat",
50
  "task": "text_classification",
@@ -63,12 +57,6 @@
63
  "acc": 0.42230347349177333,
64
  "acc_norm": 0.4227605118829982
65
  },
66
- {
67
- "name": "snli_tr",
68
- "task": "natural_language_inference",
69
- "acc": 0.336,
70
- "acc_norm": 0.3368
71
- },
72
  {
73
  "name": "sts_tr",
74
  "task": "text_classification",
@@ -87,12 +75,6 @@
87
  "acc": 0.57,
88
  "acc_norm": 0.57
89
  },
90
- {
91
- "name": "xnli_tr",
92
- "task": "natural_language_inference",
93
- "acc": 0.457429718875502,
94
- "acc_norm": 0.457429718875502
95
- },
96
  {
97
  "name": "xquad_tr",
98
  "task": "extractive_question_answering",
@@ -191,6 +173,24 @@
191
  "task": "multiple_choice",
192
  "acc": 0.5954946131243879,
193
  "acc_norm": 0.5954946131243879
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
  ]
196
  }
 
39
  "exact_match": 0.06007694584196508,
40
  "f1": 0.10654344736882515
41
  },
 
 
 
 
 
 
42
  {
43
  "name": "news_cat",
44
  "task": "text_classification",
 
57
  "acc": 0.42230347349177333,
58
  "acc_norm": 0.4227605118829982
59
  },
 
 
 
 
 
 
60
  {
61
  "name": "sts_tr",
62
  "task": "text_classification",
 
75
  "acc": 0.57,
76
  "acc_norm": 0.57
77
  },
 
 
 
 
 
 
78
  {
79
  "name": "xquad_tr",
80
  "task": "extractive_question_answering",
 
173
  "task": "multiple_choice",
174
  "acc": 0.5954946131243879,
175
  "acc_norm": 0.5954946131243879
176
+ },
177
+ {
178
+ "name": "mnli_tr",
179
+ "task": "natural_language_inference",
180
+ "acc": 0.32,
181
+ "acc_norm": 0.321
182
+ },
183
+ {
184
+ "name": "snli_tr",
185
+ "task": "natural_language_inference",
186
+ "acc": 0.325,
187
+ "acc_norm": 0.3239
188
+ },
189
+ {
190
+ "name": "xnli_tr",
191
+ "task": "natural_language_inference",
192
+ "acc": 0.32994011976047904,
193
+ "acc_norm": 0.332934131736527
194
  }
195
  ]
196
  }
results/zero-shot/mistral-7b.json CHANGED
@@ -84,20 +84,20 @@
84
  {
85
  "name": "mnli_tr",
86
  "task": "natural_language_inference",
87
- "acc": 0.3194,
88
- "acc_norm": 0.3267
89
  },
90
  {
91
  "name": "snli_tr",
92
  "task": "natural_language_inference",
93
- "acc": 0.3196,
94
- "acc_norm": 0.3201
95
  },
96
  {
97
  "name": "xnli_tr",
98
  "task": "natural_language_inference",
99
- "acc": 0.331936127744511,
100
- "acc_norm": 0.34910179640718564
101
  },
102
  {
103
  "name": "news_cat",
 
84
  {
85
  "name": "mnli_tr",
86
  "task": "natural_language_inference",
87
+ "acc": 0.3481,
88
+ "acc_norm": 0.3185
89
  },
90
  {
91
  "name": "snli_tr",
92
  "task": "natural_language_inference",
93
+ "acc": 0.3359,
94
+ "acc_norm": 0.3344
95
  },
96
  {
97
  "name": "xnli_tr",
98
  "task": "natural_language_inference",
99
+ "acc": 0.3341317365269461,
100
+ "acc_norm": 0.3189620758483034
101
  },
102
  {
103
  "name": "news_cat",
results/zero-shot/trendyol-7b.json CHANGED
@@ -111,20 +111,20 @@
111
  {
112
  "name": "mnli_tr",
113
  "task": "natural_language_inference",
114
- "acc": 0.3134,
115
- "acc_norm": 0.2942
116
  },
117
  {
118
  "name": "snli_tr",
119
  "task": "natural_language_inference",
120
- "acc": 0.3204,
121
- "acc_norm": 0.2894
122
  },
123
  {
124
  "name": "xnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.32974051896207585,
127
- "acc_norm": 0.300998003992016
128
  },
129
  {
130
  "name": "news_cat",
 
111
  {
112
  "name": "mnli_tr",
113
  "task": "natural_language_inference",
114
+ "acc": 0.3519,
115
+ "acc_norm": 0.3543
116
  },
117
  {
118
  "name": "snli_tr",
119
  "task": "natural_language_inference",
120
+ "acc": 0.3372,
121
+ "acc_norm": 0.3487
122
  },
123
  {
124
  "name": "xnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.3403193612774451,
127
+ "acc_norm": 0.35449101796407184
128
  },
129
  {
130
  "name": "news_cat",
results/zero-shot/turna.json CHANGED
@@ -111,20 +111,20 @@
111
  {
112
  "name": "mnli_tr",
113
  "task": "natural_language_inference",
114
- "acc": 0.3203,
115
- "acc_norm": 0.3159
116
  },
117
  {
118
  "name": "snli_tr",
119
  "task": "natural_language_inference",
120
- "acc": 0.3223,
121
- "acc_norm": 0.3278
122
  },
123
  {
124
  "name": "xnli_tr",
125
  "task": "natural_language_inference",
126
- "acc": 0.32974051896207585,
127
- "acc_norm": 0.3277445109780439
128
  },
129
  {
130
  "name": "news_cat",
 
111
  {
112
  "name": "mnli_tr",
113
  "task": "natural_language_inference",
114
+ "acc": 0.3494,
115
+ "acc_norm": 0.3464
116
  },
117
  {
118
  "name": "snli_tr",
119
  "task": "natural_language_inference",
120
+ "acc": 0.3381,
121
+ "acc_norm": 0.3343
122
  },
123
  {
124
  "name": "xnli_tr",
125
  "task": "natural_language_inference",
126
+ "acc": 0.3341317365269461,
127
+ "acc_norm": 0.33812375249501
128
  },
129
  {
130
  "name": "news_cat",