Hi @sunkaikai ,

Thank you very much for releasing the source code. I have tried to train and evaluate on the NYT (partial match), but I only got the best performance on NYT is: 83.21% (F1-score) in the case of partial match. Meanwhile, the reported best result on NYT by RIN model is: 87.3% (in Table 1). I wonder what should I do for getting this performance. Thanks so much!
Also, I would like to attach the file logs.txt .

epoch train_loss train_loss_er train_loss_rc dev_loss dev_loss_er dev_loss_rc P R F1
1 73.2485 3.4390 69.8095 4.2972 2.2674 2.0298 0.8443 0.7026 0.7669
new best model saved at epoch 1: 73.2485 4.2972 0.8443 0.7026 0.7669
2 3.5344 1.9053 1.6291 3.7588 1.9997 1.7591 0.8486 0.7773 0.8114
new best model saved at epoch 2: 3.5344 3.7588 0.8486 0.7773 0.8114
3 2.4777 1.3386 1.1390 3.9806 2.0999 1.8807 0.8106 0.8226 0.8165
new best model saved at epoch 3: 2.4777 3.9806 0.8106 0.8226 0.8165
4 1.7885 0.9412 0.8473 4.1865 2.3059 1.8806 0.8253 0.8176 0.8214
new best model saved at epoch 4: 1.7885 4.1865 0.8253 0.8176 0.8214
5 1.3211 0.6723 0.6488 4.5396 2.5449 1.9946 0.8277 0.8180 0.8228
new best model saved at epoch 5: 1.3211 4.5396 0.8277 0.8180 0.8228
6 1.0103 0.4998 0.5105 5.0407 2.9166 2.1241 0.8392 0.8148 0.8268
new best model saved at epoch 6: 1.0103 5.0407 0.8392 0.8148 0.8268
7 0.7966 0.3867 0.4099 5.7944 3.4451 2.3494 0.8210 0.8221 0.8215
8 0.6689 0.3082 0.3608 5.9889 3.4514 2.5374 0.8069 0.8313 0.8189
9 0.5905 0.2657 0.3248 6.5597 3.8226 2.7371 0.7920 0.8497 0.8198
10 0.5239 0.2350 0.2889 6.0840 3.4209 2.6631 0.8172 0.8354 0.8262
11 0.4712 0.2044 0.2668 6.1135 3.5208 2.5926 0.8488 0.8160 0.8321
new best model saved at epoch 11: 0.4712 6.1135 0.8488 0.8160 0.8321
12 0.4310 0.1831 0.2479 6.5186 3.8178 2.7008 0.8271 0.8260 0.8266
13 0.4003 0.1692 0.2311 6.3287 3.6748 2.6539 0.8086 0.8508 0.8292
14 0.3760 0.1589 0.2171 7.1753 4.0607 3.1147 0.8109 0.8297 0.8202
15 0.3468 0.1429 0.2039 6.8977 4.0689 2.8288 0.8203 0.8353 0.8277
16 0.3426 0.1379 0.2047 7.5494 4.3998 3.1496 0.8261 0.8231 0.8246
17 0.3212 0.1318 0.1894 7.4385 4.3242 3.1143 0.8026 0.8457 0.8236
18 0.2987 0.1161 0.1825 7.8317 4.3870 3.4447 0.8174 0.8224 0.8199
19 0.3014 0.1221 0.1793 7.1426 4.0276 3.1150 0.8094 0.8374 0.8232
20 0.2867 0.1129 0.1737 7.4272 4.2601 3.1672 0.8333 0.8196 0.8264
21 0.2729 0.1090 0.1639 7.3648 4.0702 3.2946 0.8269 0.8195 0.8232
22 0.2703 0.1062 0.1641 7.2367 4.1408 3.0958 0.8467 0.8111 0.8285
23 0.2564 0.0998 0.1566 8.1523 4.6444 3.5079 0.8174 0.8248 0.8211
24 0.2298 0.0887 0.1412 7.9464 4.5360 3.4104 0.8135 0.8271 0.8202
25 0.2450 0.0950 0.1500 7.3765 4.1272 3.2493 0.8188 0.8253 0.8220
26 0.2355 0.0905 0.1450 7.4168 4.3299 3.0869 0.8139 0.8295 0.8216
27 0.2237 0.0869 0.1368 7.8651 4.5665 3.2986 0.8275 0.8228 0.8252
28 0.2149 0.0828 0.1321 7.5777 4.1679 3.4098 0.8349 0.8289 0.8319
29 0.2124 0.0796 0.1329 8.1536 4.6870 3.4666 0.8180 0.8178 0.8179
30 0.2151 0.0788 0.1362 7.4739 4.3265 3.1474 0.8316 0.8208 0.8262
31 0.2076 0.0782 0.1294 8.1315 4.6649 3.4666 0.8311 0.8170 0.8240
32 0.2001 0.0716 0.1284 7.8508 4.4605 3.3903 0.8427 0.8091 0.8256
33 0.2030 0.0755 0.1275 7.7402 4.3821 3.3582 0.8404 0.7942 0.8167
34 0.1971 0.0735 0.1236 8.1181 4.6893 3.4288 0.8269 0.8213 0.8241
35 0.1926 0.0713 0.1213 8.9690 4.9311 4.0379 0.8262 0.8091 0.8176
36 0.1890 0.0676 0.1215 8.4963 4.8067 3.6896 0.8059 0.8330 0.8193
37 0.1917 0.0681 0.1235 8.3412 4.6584 3.6828 0.8324 0.7984 0.8150
38 0.1897 0.0686 0.1211 8.6424 4.8544 3.7880 0.8371 0.8059 0.8212
39 0.1779 0.0648 0.1131 8.4247 4.6384 3.7863 0.8403 0.8033 0.8214
40 0.1863 0.0669 0.1194 8.2358 4.6130 3.6228 0.8288 0.8109 0.8197