Spaces:
Running
Running
Upload folder using huggingface_hub
Browse files- results/voyage-3-large.json +546 -0
results/voyage-3-large.json
ADDED
|
@@ -0,0 +1,546 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"task_id": "conversation_similarity",
|
| 4 |
+
"task_name": "Conversation Similarity",
|
| 5 |
+
"model_name": "voyage-3-large",
|
| 6 |
+
"benchmark_version": "0.1.0",
|
| 7 |
+
"dataset_revision": "",
|
| 8 |
+
"mrr_at_10": 0.1140340689121177,
|
| 9 |
+
"map_at_10": 0.05175022583559168,
|
| 10 |
+
"recall_at_1": 0.0,
|
| 11 |
+
"recall_at_5": 0.23170731707317074,
|
| 12 |
+
"recall_at_10": 0.45121951219512196,
|
| 13 |
+
"ndcg_at_10": 0.10921881438171957,
|
| 14 |
+
"num_queries": 82,
|
| 15 |
+
"num_corpus_docs": 1595,
|
| 16 |
+
"confidence_intervals": {
|
| 17 |
+
"MRR@10": [
|
| 18 |
+
0.07998923248161052,
|
| 19 |
+
0.14816056910569103
|
| 20 |
+
],
|
| 21 |
+
"MAP@10": [
|
| 22 |
+
0.033713221060782034,
|
| 23 |
+
0.07145680087753259
|
| 24 |
+
],
|
| 25 |
+
"R@1": [
|
| 26 |
+
0.0,
|
| 27 |
+
0.0
|
| 28 |
+
],
|
| 29 |
+
"R@5": [
|
| 30 |
+
0.14634146341463414,
|
| 31 |
+
0.31737804878048753
|
| 32 |
+
],
|
| 33 |
+
"R@10": [
|
| 34 |
+
0.34146341463414637,
|
| 35 |
+
0.5490853658536583
|
| 36 |
+
],
|
| 37 |
+
"NDCG@10": [
|
| 38 |
+
0.07815159631767281,
|
| 39 |
+
0.13875558398642276
|
| 40 |
+
]
|
| 41 |
+
},
|
| 42 |
+
"per_difficulty": {
|
| 43 |
+
"unknown": {
|
| 44 |
+
"MRR@10": 0.1140340689121177,
|
| 45 |
+
"MAP@10": 0.05175022583559168,
|
| 46 |
+
"R@1": 0.0,
|
| 47 |
+
"R@5": 0.23170731707317074,
|
| 48 |
+
"R@10": 0.45121951219512196,
|
| 49 |
+
"NDCG@10": 0.10921881438171957,
|
| 50 |
+
"count": 82
|
| 51 |
+
}
|
| 52 |
+
},
|
| 53 |
+
"hard_negative_metrics": {},
|
| 54 |
+
"robustness_score": null,
|
| 55 |
+
"metric_std_devs": {},
|
| 56 |
+
"mean_query_stability": null,
|
| 57 |
+
"n_paraphrases": 0
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"task_id": "cross_channel",
|
| 61 |
+
"task_name": "Cross Channel",
|
| 62 |
+
"model_name": "voyage-3-large",
|
| 63 |
+
"benchmark_version": "0.1.0",
|
| 64 |
+
"dataset_revision": "",
|
| 65 |
+
"mrr_at_10": 0.3200418314136726,
|
| 66 |
+
"map_at_10": 0.15571194582927436,
|
| 67 |
+
"recall_at_1": 0.17509025270758122,
|
| 68 |
+
"recall_at_5": 0.5126353790613718,
|
| 69 |
+
"recall_at_10": 0.7220216606498195,
|
| 70 |
+
"ndcg_at_10": 0.2551008842387347,
|
| 71 |
+
"num_queries": 554,
|
| 72 |
+
"num_corpus_docs": 1595,
|
| 73 |
+
"confidence_intervals": {
|
| 74 |
+
"MRR@10": [
|
| 75 |
+
0.2915454164517792,
|
| 76 |
+
0.3524498237923328
|
| 77 |
+
],
|
| 78 |
+
"MAP@10": [
|
| 79 |
+
0.1409206671537448,
|
| 80 |
+
0.1720265567995148
|
| 81 |
+
],
|
| 82 |
+
"R@1": [
|
| 83 |
+
0.1444043321299639,
|
| 84 |
+
0.2075812274368231
|
| 85 |
+
],
|
| 86 |
+
"R@5": [
|
| 87 |
+
0.4729241877256318,
|
| 88 |
+
0.5577617328519856
|
| 89 |
+
],
|
| 90 |
+
"R@10": [
|
| 91 |
+
0.6841155234657039,
|
| 92 |
+
0.759927797833935
|
| 93 |
+
],
|
| 94 |
+
"NDCG@10": [
|
| 95 |
+
0.2363556281226957,
|
| 96 |
+
0.27561158230539357
|
| 97 |
+
]
|
| 98 |
+
},
|
| 99 |
+
"per_difficulty": {
|
| 100 |
+
"easy": {
|
| 101 |
+
"MRR@10": 0.2559862976529643,
|
| 102 |
+
"MAP@10": 0.11582214082214082,
|
| 103 |
+
"R@1": 0.10256410256410256,
|
| 104 |
+
"R@5": 0.46153846153846156,
|
| 105 |
+
"R@10": 0.7264957264957265,
|
| 106 |
+
"NDCG@10": 0.21392533137141628,
|
| 107 |
+
"count": 117
|
| 108 |
+
},
|
| 109 |
+
"hard": {
|
| 110 |
+
"MRR@10": 0.3544979873927242,
|
| 111 |
+
"MAP@10": 0.1653557821320979,
|
| 112 |
+
"R@1": 0.22009569377990432,
|
| 113 |
+
"R@5": 0.5215311004784688,
|
| 114 |
+
"R@10": 0.6985645933014354,
|
| 115 |
+
"NDCG@10": 0.2625001923284304,
|
| 116 |
+
"count": 209
|
| 117 |
+
},
|
| 118 |
+
"medium": {
|
| 119 |
+
"MRR@10": 0.3213276246170983,
|
| 120 |
+
"MAP@10": 0.16734153091060983,
|
| 121 |
+
"R@1": 0.17105263157894737,
|
| 122 |
+
"R@5": 0.5307017543859649,
|
| 123 |
+
"R@10": 0.7412280701754386,
|
| 124 |
+
"NDCG@10": 0.2694477451805324,
|
| 125 |
+
"count": 228
|
| 126 |
+
}
|
| 127 |
+
},
|
| 128 |
+
"hard_negative_metrics": {
|
| 129 |
+
"hn_mean_rank": 9.653429602888087,
|
| 130 |
+
"hn_above_relevant_rate": 0.12815884476534295,
|
| 131 |
+
"hn_query_count": 554
|
| 132 |
+
},
|
| 133 |
+
"robustness_score": null,
|
| 134 |
+
"metric_std_devs": {},
|
| 135 |
+
"mean_query_stability": null,
|
| 136 |
+
"n_paraphrases": 0
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"task_id": "cross_platform_transfer",
|
| 140 |
+
"task_name": "Cross-Platform Transfer (slack)",
|
| 141 |
+
"model_name": "voyage-3-large",
|
| 142 |
+
"benchmark_version": "0.1.0",
|
| 143 |
+
"dataset_revision": "",
|
| 144 |
+
"mrr_at_10": 0.8568057311420142,
|
| 145 |
+
"map_at_10": 0.8568057311420142,
|
| 146 |
+
"recall_at_1": 0.7964601769911505,
|
| 147 |
+
"recall_at_5": 0.9292035398230089,
|
| 148 |
+
"recall_at_10": 0.9734513274336283,
|
| 149 |
+
"ndcg_at_10": 0.8852294883180861,
|
| 150 |
+
"num_queries": 113,
|
| 151 |
+
"num_corpus_docs": 1232,
|
| 152 |
+
"confidence_intervals": {
|
| 153 |
+
"MRR@10": [
|
| 154 |
+
0.8005193847450484,
|
| 155 |
+
0.9074083438685208
|
| 156 |
+
],
|
| 157 |
+
"MAP@10": [
|
| 158 |
+
0.8005193847450484,
|
| 159 |
+
0.9074083438685208
|
| 160 |
+
],
|
| 161 |
+
"R@1": [
|
| 162 |
+
0.7168141592920354,
|
| 163 |
+
0.8672566371681416
|
| 164 |
+
],
|
| 165 |
+
"R@5": [
|
| 166 |
+
0.8849557522123894,
|
| 167 |
+
0.9734513274336283
|
| 168 |
+
],
|
| 169 |
+
"R@10": [
|
| 170 |
+
0.9380530973451328,
|
| 171 |
+
1.0
|
| 172 |
+
],
|
| 173 |
+
"NDCG@10": [
|
| 174 |
+
0.8377052536121854,
|
| 175 |
+
0.9281102743762883
|
| 176 |
+
]
|
| 177 |
+
},
|
| 178 |
+
"per_difficulty": {
|
| 179 |
+
"unknown": {
|
| 180 |
+
"MRR@10": 0.8568057311420142,
|
| 181 |
+
"MAP@10": 0.8568057311420142,
|
| 182 |
+
"R@1": 0.7964601769911505,
|
| 183 |
+
"R@5": 0.9292035398230089,
|
| 184 |
+
"R@10": 0.9734513274336283,
|
| 185 |
+
"NDCG@10": 0.8852294883180861,
|
| 186 |
+
"count": 113
|
| 187 |
+
}
|
| 188 |
+
},
|
| 189 |
+
"hard_negative_metrics": {},
|
| 190 |
+
"robustness_score": null,
|
| 191 |
+
"metric_std_devs": {},
|
| 192 |
+
"mean_query_stability": null,
|
| 193 |
+
"n_paraphrases": 0
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"task_id": "response_retrieval",
|
| 197 |
+
"task_name": "Response Retrieval",
|
| 198 |
+
"model_name": "voyage-3-large",
|
| 199 |
+
"benchmark_version": "0.1.0",
|
| 200 |
+
"dataset_revision": "",
|
| 201 |
+
"mrr_at_10": 0.48689126984126985,
|
| 202 |
+
"map_at_10": 0.48689126984126985,
|
| 203 |
+
"recall_at_1": 0.416,
|
| 204 |
+
"recall_at_5": 0.574,
|
| 205 |
+
"recall_at_10": 0.64,
|
| 206 |
+
"ndcg_at_10": 0.5235468640787492,
|
| 207 |
+
"num_queries": 500,
|
| 208 |
+
"num_corpus_docs": 1173,
|
| 209 |
+
"confidence_intervals": {
|
| 210 |
+
"MRR@10": [
|
| 211 |
+
0.4463512301587302,
|
| 212 |
+
0.5278171428571428
|
| 213 |
+
],
|
| 214 |
+
"MAP@10": [
|
| 215 |
+
0.4463512301587302,
|
| 216 |
+
0.5278171428571428
|
| 217 |
+
],
|
| 218 |
+
"R@1": [
|
| 219 |
+
0.37395000000000006,
|
| 220 |
+
0.462
|
| 221 |
+
],
|
| 222 |
+
"R@5": [
|
| 223 |
+
0.53,
|
| 224 |
+
0.618
|
| 225 |
+
],
|
| 226 |
+
"R@10": [
|
| 227 |
+
0.596,
|
| 228 |
+
0.68005
|
| 229 |
+
],
|
| 230 |
+
"NDCG@10": [
|
| 231 |
+
0.48266746398429305,
|
| 232 |
+
0.5630537660518903
|
| 233 |
+
]
|
| 234 |
+
},
|
| 235 |
+
"per_difficulty": {
|
| 236 |
+
"unknown": {
|
| 237 |
+
"MRR@10": 0.48689126984126985,
|
| 238 |
+
"MAP@10": 0.48689126984126985,
|
| 239 |
+
"R@1": 0.416,
|
| 240 |
+
"R@5": 0.574,
|
| 241 |
+
"R@10": 0.64,
|
| 242 |
+
"NDCG@10": 0.5235468640787492,
|
| 243 |
+
"count": 500
|
| 244 |
+
}
|
| 245 |
+
},
|
| 246 |
+
"hard_negative_metrics": {},
|
| 247 |
+
"robustness_score": null,
|
| 248 |
+
"metric_std_devs": {},
|
| 249 |
+
"mean_query_stability": null,
|
| 250 |
+
"n_paraphrases": 0
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"task_id": "specific_detail",
|
| 254 |
+
"task_name": "Specific Detail",
|
| 255 |
+
"model_name": "voyage-3-large",
|
| 256 |
+
"benchmark_version": "0.1.0",
|
| 257 |
+
"dataset_revision": "",
|
| 258 |
+
"mrr_at_10": 0.3294289196482981,
|
| 259 |
+
"map_at_10": 0.3294289196482981,
|
| 260 |
+
"recall_at_1": 0.1773308957952468,
|
| 261 |
+
"recall_at_5": 0.5301645338208409,
|
| 262 |
+
"recall_at_10": 0.7294332723948812,
|
| 263 |
+
"ndcg_at_10": 0.42377023479275683,
|
| 264 |
+
"num_queries": 547,
|
| 265 |
+
"num_corpus_docs": 1595,
|
| 266 |
+
"confidence_intervals": {
|
| 267 |
+
"MRR@10": [
|
| 268 |
+
0.29899881387655614,
|
| 269 |
+
0.3577598045616784
|
| 270 |
+
],
|
| 271 |
+
"MAP@10": [
|
| 272 |
+
0.29899881387655614,
|
| 273 |
+
0.3577598045616784
|
| 274 |
+
],
|
| 275 |
+
"R@1": [
|
| 276 |
+
0.14620658135283368,
|
| 277 |
+
0.21023765996343693
|
| 278 |
+
],
|
| 279 |
+
"R@5": [
|
| 280 |
+
0.4880712979890311,
|
| 281 |
+
0.5685557586837294
|
| 282 |
+
],
|
| 283 |
+
"R@10": [
|
| 284 |
+
0.6892138939670932,
|
| 285 |
+
0.7659963436928702
|
| 286 |
+
],
|
| 287 |
+
"NDCG@10": [
|
| 288 |
+
0.3942246959871027,
|
| 289 |
+
0.4513162163232951
|
| 290 |
+
]
|
| 291 |
+
},
|
| 292 |
+
"per_difficulty": {
|
| 293 |
+
"easy": {
|
| 294 |
+
"MRR@10": 0.3287136867325547,
|
| 295 |
+
"MAP@10": 0.3287136867325547,
|
| 296 |
+
"R@1": 0.16981132075471697,
|
| 297 |
+
"R@5": 0.5566037735849056,
|
| 298 |
+
"R@10": 0.8018867924528302,
|
| 299 |
+
"NDCG@10": 0.4395466947913532,
|
| 300 |
+
"count": 106
|
| 301 |
+
},
|
| 302 |
+
"hard": {
|
| 303 |
+
"MRR@10": 0.318499373433584,
|
| 304 |
+
"MAP@10": 0.318499373433584,
|
| 305 |
+
"R@1": 0.17543859649122806,
|
| 306 |
+
"R@5": 0.49122807017543857,
|
| 307 |
+
"R@10": 0.6842105263157895,
|
| 308 |
+
"NDCG@10": 0.4045863919363865,
|
| 309 |
+
"count": 228
|
| 310 |
+
},
|
| 311 |
+
"medium": {
|
| 312 |
+
"MRR@10": 0.34148408972352634,
|
| 313 |
+
"MAP@10": 0.34148408972352634,
|
| 314 |
+
"R@1": 0.18309859154929578,
|
| 315 |
+
"R@5": 0.5586854460093896,
|
| 316 |
+
"R@10": 0.7417840375586855,
|
| 317 |
+
"NDCG@10": 0.43645385644252777,
|
| 318 |
+
"count": 213
|
| 319 |
+
}
|
| 320 |
+
},
|
| 321 |
+
"hard_negative_metrics": {
|
| 322 |
+
"hn_mean_rank": 7.595978062157221,
|
| 323 |
+
"hn_above_relevant_rate": 0.2979890310786106,
|
| 324 |
+
"hn_query_count": 547
|
| 325 |
+
},
|
| 326 |
+
"robustness_score": null,
|
| 327 |
+
"metric_std_devs": {},
|
| 328 |
+
"mean_query_stability": null,
|
| 329 |
+
"n_paraphrases": 0
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"task_id": "thread_discrimination",
|
| 333 |
+
"task_name": "Thread Discrimination",
|
| 334 |
+
"model_name": "voyage-3-large",
|
| 335 |
+
"benchmark_version": "0.1.0",
|
| 336 |
+
"dataset_revision": "",
|
| 337 |
+
"mrr_at_10": 0.5443265755146942,
|
| 338 |
+
"map_at_10": 0.5443265755146942,
|
| 339 |
+
"recall_at_1": 0.3782178217821782,
|
| 340 |
+
"recall_at_5": 0.7683168316831683,
|
| 341 |
+
"recall_at_10": 0.9188118811881189,
|
| 342 |
+
"ndcg_at_10": 0.6338301198377989,
|
| 343 |
+
"num_queries": 505,
|
| 344 |
+
"num_corpus_docs": 1595,
|
| 345 |
+
"confidence_intervals": {
|
| 346 |
+
"MRR@10": [
|
| 347 |
+
0.5099766619519094,
|
| 348 |
+
0.5764179828697155
|
| 349 |
+
],
|
| 350 |
+
"MAP@10": [
|
| 351 |
+
0.5099766619519094,
|
| 352 |
+
0.5764179828697155
|
| 353 |
+
],
|
| 354 |
+
"R@1": [
|
| 355 |
+
0.33663366336633666,
|
| 356 |
+
0.4178217821782178
|
| 357 |
+
],
|
| 358 |
+
"R@5": [
|
| 359 |
+
0.7287128712871287,
|
| 360 |
+
0.805940594059406
|
| 361 |
+
],
|
| 362 |
+
"R@10": [
|
| 363 |
+
0.8950495049504951,
|
| 364 |
+
0.9405940594059405
|
| 365 |
+
],
|
| 366 |
+
"NDCG@10": [
|
| 367 |
+
0.6042357834066427,
|
| 368 |
+
0.6608133396278401
|
| 369 |
+
]
|
| 370 |
+
},
|
| 371 |
+
"per_difficulty": {
|
| 372 |
+
"easy": {
|
| 373 |
+
"MRR@10": 0.5475919913419914,
|
| 374 |
+
"MAP@10": 0.5475919913419914,
|
| 375 |
+
"R@1": 0.3409090909090909,
|
| 376 |
+
"R@5": 0.8181818181818182,
|
| 377 |
+
"R@10": 0.9772727272727273,
|
| 378 |
+
"NDCG@10": 0.6510224048778747,
|
| 379 |
+
"count": 44
|
| 380 |
+
},
|
| 381 |
+
"hard": {
|
| 382 |
+
"MRR@10": 0.5082574895714826,
|
| 383 |
+
"MAP@10": 0.5082574895714826,
|
| 384 |
+
"R@1": 0.3378839590443686,
|
| 385 |
+
"R@5": 0.7406143344709898,
|
| 386 |
+
"R@10": 0.9112627986348123,
|
| 387 |
+
"NDCG@10": 0.6042707869145505,
|
| 388 |
+
"count": 293
|
| 389 |
+
},
|
| 390 |
+
"medium": {
|
| 391 |
+
"MRR@10": 0.6063775510204081,
|
| 392 |
+
"MAP@10": 0.6063775510204081,
|
| 393 |
+
"R@1": 0.4583333333333333,
|
| 394 |
+
"R@5": 0.8035714285714286,
|
| 395 |
+
"R@10": 0.9166666666666666,
|
| 396 |
+
"NDCG@10": 0.6808802627232063,
|
| 397 |
+
"count": 168
|
| 398 |
+
}
|
| 399 |
+
},
|
| 400 |
+
"hard_negative_metrics": {
|
| 401 |
+
"hn_mean_rank": 4.9485148514851485,
|
| 402 |
+
"hn_above_relevant_rate": 0.3603960396039604,
|
| 403 |
+
"hn_query_count": 505
|
| 404 |
+
},
|
| 405 |
+
"robustness_score": null,
|
| 406 |
+
"metric_std_devs": {},
|
| 407 |
+
"mean_query_stability": null,
|
| 408 |
+
"n_paraphrases": 0
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"task_id": "thread_retrieval",
|
| 412 |
+
"task_name": "Thread Retrieval",
|
| 413 |
+
"model_name": "voyage-3-large",
|
| 414 |
+
"benchmark_version": "0.1.0",
|
| 415 |
+
"dataset_revision": "",
|
| 416 |
+
"mrr_at_10": 0.516886507936508,
|
| 417 |
+
"map_at_10": 0.516886507936508,
|
| 418 |
+
"recall_at_1": 0.45,
|
| 419 |
+
"recall_at_5": 0.614,
|
| 420 |
+
"recall_at_10": 0.654,
|
| 421 |
+
"ndcg_at_10": 0.5500967749445125,
|
| 422 |
+
"num_queries": 500,
|
| 423 |
+
"num_corpus_docs": 1232,
|
| 424 |
+
"confidence_intervals": {
|
| 425 |
+
"MRR@10": [
|
| 426 |
+
0.47520793650793647,
|
| 427 |
+
0.5578661706349206
|
| 428 |
+
],
|
| 429 |
+
"MAP@10": [
|
| 430 |
+
0.47520793650793647,
|
| 431 |
+
0.5578661706349206
|
| 432 |
+
],
|
| 433 |
+
"R@1": [
|
| 434 |
+
0.406,
|
| 435 |
+
0.494
|
| 436 |
+
],
|
| 437 |
+
"R@5": [
|
| 438 |
+
0.57,
|
| 439 |
+
0.66
|
| 440 |
+
],
|
| 441 |
+
"R@10": [
|
| 442 |
+
0.612,
|
| 443 |
+
0.698
|
| 444 |
+
],
|
| 445 |
+
"NDCG@10": [
|
| 446 |
+
0.5086154307321877,
|
| 447 |
+
0.5916009947937753
|
| 448 |
+
]
|
| 449 |
+
},
|
| 450 |
+
"per_difficulty": {
|
| 451 |
+
"unknown": {
|
| 452 |
+
"MRR@10": 0.516886507936508,
|
| 453 |
+
"MAP@10": 0.516886507936508,
|
| 454 |
+
"R@1": 0.45,
|
| 455 |
+
"R@5": 0.614,
|
| 456 |
+
"R@10": 0.654,
|
| 457 |
+
"NDCG@10": 0.5500967749445125,
|
| 458 |
+
"count": 500
|
| 459 |
+
}
|
| 460 |
+
},
|
| 461 |
+
"hard_negative_metrics": {},
|
| 462 |
+
"robustness_score": null,
|
| 463 |
+
"metric_std_devs": {},
|
| 464 |
+
"mean_query_stability": null,
|
| 465 |
+
"n_paraphrases": 0
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"task_id": "topic_retrieval",
|
| 469 |
+
"task_name": "Topic Retrieval",
|
| 470 |
+
"model_name": "voyage-3-large",
|
| 471 |
+
"benchmark_version": "0.1.0",
|
| 472 |
+
"dataset_revision": "",
|
| 473 |
+
"mrr_at_10": 0.47772130944645563,
|
| 474 |
+
"map_at_10": 0.475276926885114,
|
| 475 |
+
"recall_at_1": 0.3079922027290448,
|
| 476 |
+
"recall_at_5": 0.6998050682261209,
|
| 477 |
+
"recall_at_10": 0.9103313840155945,
|
| 478 |
+
"ndcg_at_10": 0.5786813479948444,
|
| 479 |
+
"num_queries": 513,
|
| 480 |
+
"num_corpus_docs": 1595,
|
| 481 |
+
"confidence_intervals": {
|
| 482 |
+
"MRR@10": [
|
| 483 |
+
0.44517704368947053,
|
| 484 |
+
0.5107191590086326
|
| 485 |
+
],
|
| 486 |
+
"MAP@10": [
|
| 487 |
+
0.44275772378477063,
|
| 488 |
+
0.5082153299916458
|
| 489 |
+
],
|
| 490 |
+
"R@1": [
|
| 491 |
+
0.26900584795321636,
|
| 492 |
+
0.3489278752436647
|
| 493 |
+
],
|
| 494 |
+
"R@5": [
|
| 495 |
+
0.6607699805068227,
|
| 496 |
+
0.7387914230019493
|
| 497 |
+
],
|
| 498 |
+
"R@10": [
|
| 499 |
+
0.8849415204678363,
|
| 500 |
+
0.9318226120857699
|
| 501 |
+
],
|
| 502 |
+
"NDCG@10": [
|
| 503 |
+
0.5509874540825015,
|
| 504 |
+
0.6076346782730433
|
| 505 |
+
]
|
| 506 |
+
},
|
| 507 |
+
"per_difficulty": {
|
| 508 |
+
"easy": {
|
| 509 |
+
"MRR@10": 0.4537283130306386,
|
| 510 |
+
"MAP@10": 0.4537283130306386,
|
| 511 |
+
"R@1": 0.2713178294573643,
|
| 512 |
+
"R@5": 0.7286821705426356,
|
| 513 |
+
"R@10": 0.9147286821705426,
|
| 514 |
+
"NDCG@10": 0.5633487416393298,
|
| 515 |
+
"count": 129
|
| 516 |
+
},
|
| 517 |
+
"hard": {
|
| 518 |
+
"MRR@10": 0.5193487394957983,
|
| 519 |
+
"MAP@10": 0.5120424836601307,
|
| 520 |
+
"R@1": 0.35294117647058826,
|
| 521 |
+
"R@5": 0.7176470588235294,
|
| 522 |
+
"R@10": 0.9117647058823529,
|
| 523 |
+
"NDCG@10": 0.6090460583949658,
|
| 524 |
+
"count": 170
|
| 525 |
+
},
|
| 526 |
+
"medium": {
|
| 527 |
+
"MRR@10": 0.4591158581812787,
|
| 528 |
+
"MAP@10": 0.4590602284527518,
|
| 529 |
+
"R@1": 0.29439252336448596,
|
| 530 |
+
"R@5": 0.6682242990654206,
|
| 531 |
+
"R@10": 0.9065420560747663,
|
| 532 |
+
"NDCG@10": 0.5638024015081188,
|
| 533 |
+
"count": 214
|
| 534 |
+
}
|
| 535 |
+
},
|
| 536 |
+
"hard_negative_metrics": {
|
| 537 |
+
"hn_mean_rank": 7.105263157894737,
|
| 538 |
+
"hn_above_relevant_rate": 0.27485380116959063,
|
| 539 |
+
"hn_query_count": 513
|
| 540 |
+
},
|
| 541 |
+
"robustness_score": null,
|
| 542 |
+
"metric_std_devs": {},
|
| 543 |
+
"mean_query_stability": null,
|
| 544 |
+
"n_paraphrases": 0
|
| 545 |
+
}
|
| 546 |
+
]
|