File size: 61,753 Bytes
6fa4bc9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 |
{
"paper_id": "2020",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T07:13:13.525984Z"
},
"title": "Counterfactually-Augmented SNLI Training Data Does Not Yield Better Generalization Than Unaugmented Data",
"authors": [
{
"first": "William",
"middle": [],
"last": "Huang",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "New York University",
"location": {}
},
"email": "will.huang@nyu.edu"
},
{
"first": "Haokun",
"middle": [],
"last": "Liu",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "New York University",
"location": {}
},
"email": "haokunliu@nyu.edu"
},
{
"first": "Samuel",
"middle": [
"R"
],
"last": "Bowman",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "New York University",
"location": {}
},
"email": "bowman@nyu.edu"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "A growing body of work shows that models exploit annotation artifacts to achieve state-ofthe-art performance on standard crowdsourced benchmarks-datasets collected from crowdworkers to create an evaluation task-while still failing on out-of-domain examples for the same task. Recent work has explored the use of counterfactually-augmented data-data built by minimally editing a set of seed examples to yield counterfactual labels-to augment training data associated with these benchmarks and build more robust classifiers that generalize better. However, Khashabi et al. (2020) find that this type of augmentation yields little benefit on reading comprehension tasks when controlling for dataset size and cost of collection. We build upon this work by using English natural language inference data to test model generalization and robustness and find that models trained on a counterfactuallyaugmented SNLI dataset do not generalize better than unaugmented datasets of similar size and that counterfactual augmentation can hurt performance, yielding models that are less robust to challenge examples. Counterfactual augmentation of natural language understanding data through standard crowdsourcing techniques does not appear to be an effective way of collecting training data and further innovation is required to make this general line of work viable.",
"pdf_parse": {
"paper_id": "2020",
"_pdf_hash": "",
"abstract": [
{
"text": "A growing body of work shows that models exploit annotation artifacts to achieve state-ofthe-art performance on standard crowdsourced benchmarks-datasets collected from crowdworkers to create an evaluation task-while still failing on out-of-domain examples for the same task. Recent work has explored the use of counterfactually-augmented data-data built by minimally editing a set of seed examples to yield counterfactual labels-to augment training data associated with these benchmarks and build more robust classifiers that generalize better. However, Khashabi et al. (2020) find that this type of augmentation yields little benefit on reading comprehension tasks when controlling for dataset size and cost of collection. We build upon this work by using English natural language inference data to test model generalization and robustness and find that models trained on a counterfactuallyaugmented SNLI dataset do not generalize better than unaugmented datasets of similar size and that counterfactual augmentation can hurt performance, yielding models that are less robust to challenge examples. Counterfactual augmentation of natural language understanding data through standard crowdsourcing techniques does not appear to be an effective way of collecting training data and further innovation is required to make this general line of work viable.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "While standard crowdsourced benchmarks have helped create significant progress within natural language processing (NLP), a growing body of evidence shows the existence of exploitable annotation artifacts in these datasets (Gururangan et al., 2018; Poliak et al., 2018; Tsuchiya, 2018) and that models can use artifacts to achieve state-of-the-art performance on these benchmarks (McCoy et al., 2019; Naik et al., 2018) . The existence of these artifacts makes it difficult to predict out-of-domain generalization and creates uncertainty around the abilities these tasks are designed to test.",
"cite_spans": [
{
"start": 242,
"end": 247,
"text": "2018;",
"ref_id": "BIBREF14"
},
{
"start": 248,
"end": 268,
"text": "Poliak et al., 2018;",
"ref_id": "BIBREF13"
},
{
"start": 269,
"end": 284,
"text": "Tsuchiya, 2018)",
"ref_id": "BIBREF14"
},
{
"start": 379,
"end": 399,
"text": "(McCoy et al., 2019;",
"ref_id": "BIBREF8"
},
{
"start": 400,
"end": 418,
"text": "Naik et al., 2018)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Recent work has explored using counterfactually-augmented datasets to address annotation artifacts with the intent to build more robust classifiers (Kaushik et al., 2020; Khashabi et al., 2020) . These datasets are collected by first sampling a set of seed examples and then creating new examples by minimally editing the seed examples to yield counterfactual labels. This type of data collection has been found to mitigate the presence of artifacts in SNLI (Bowman et al., 2015) and is presented as a way to \"elucidate the difference that makes a difference\" (Kaushik et al., 2020) . Further, Khashabi et al. (2020) present this as an efficient method to collect training data yielding models that are \"more robust to minor variations and generalize better\" (Khashabi et al., 2020) . However, they also find that unaugmented datasets yield better performance than datasets with 50-50 original-to-augmented data when controlling for training set size and annotation cost.",
"cite_spans": [
{
"start": 148,
"end": 170,
"text": "(Kaushik et al., 2020;",
"ref_id": "BIBREF5"
},
{
"start": 171,
"end": 193,
"text": "Khashabi et al., 2020)",
"ref_id": "BIBREF6"
},
{
"start": 453,
"end": 479,
"text": "SNLI (Bowman et al., 2015)",
"ref_id": null
},
{
"start": 560,
"end": 582,
"text": "(Kaushik et al., 2020)",
"ref_id": "BIBREF5"
},
{
"start": 594,
"end": 616,
"text": "Khashabi et al. (2020)",
"ref_id": "BIBREF6"
},
{
"start": 759,
"end": 782,
"text": "(Khashabi et al., 2020)",
"ref_id": "BIBREF6"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "In our work, we further study whether training with counterfactually-augmented data collected through standard crowdsourcing methods yields models with better generalization and robustness by focusing on the domain of natural language inference (NLI): the task of inferring whether a hypothesis is true given a true premise. We train and compare RoBERTa (Liu et al., 2019) trained on three different datasets: (1) the counterfactuallyaugmented natural language inference (CNLI) training set of 8.3k seed and augmented SNLI examples from Kaushik et al. (2020) , (2) a subsampled set of 8.3k unaugmented SNLI examples to control for size, and (3) the 1.7k CNLI seed examples originally sampled from SNLI. We then compare model performances on MNLI (Williams et al., 2018) -a dataset for the same task with examples out-of-domain to SNLI-and two diagnostic sets (Naik et al., 2018; Wang et al., 2019a) .",
"cite_spans": [
{
"start": 354,
"end": 372,
"text": "(Liu et al., 2019)",
"ref_id": null
},
{
"start": 537,
"end": 558,
"text": "Kaushik et al. (2020)",
"ref_id": "BIBREF5"
},
{
"start": 746,
"end": 769,
"text": "(Williams et al., 2018)",
"ref_id": "BIBREF17"
},
{
"start": 859,
"end": 878,
"text": "(Naik et al., 2018;",
"ref_id": null
},
{
"start": 879,
"end": 898,
"text": "Wang et al., 2019a)",
"ref_id": "BIBREF15"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "We find that RoBERTa trained on CNLI yields similar performance on out-of-domain MNLI examples when compared to the unaugmented subsampled SNLI training set and that including counterfactually-augmented examples to the CNLI seed set improves generalization. Further, we find that the improvement over seed examples correspond to an increase in n-grams from the addition of augmented examples, roughly doubling the number of 4-grams, and may be a result of improved lexical diversity from a larger training set. While we see similar trends in most of our diagnostic evaluations, we also find evidence that including augmented examples can yield worse performance than only training with seed examples.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "While there is evidence of the benefits of using this type of data for model evaluation (Gardner et al., 2020) , we find that using counterfactuallyaugmented data for training yields less robust models. We argue that further innovation is required to effectively crowdsource counterfactuallyaugmented natural language understanding (NLU) data for training more robust models with better generalization.",
"cite_spans": [
{
"start": 88,
"end": 110,
"text": "(Gardner et al., 2020)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Recent works show that several NLI benchmark datasets contain exploitable annotation artifacts. Several studies (Poliak et al., 2018; Gururangan et al., 2018; Tsuchiya, 2018) show that models trained on hypothesis-only examples manage to perform as much as 35 points higher than chance. Gururangan et al. (2018) also find negation words such as no or never are strongly associated with contradiction predictions. Other works (Naik et al., 2018; McCoy et al., 2019) find that models can exploit premise-hypothesis word overlap to achieve state-of-the-art performance on benchmarks by using associations of high overlap with entailment predictions and low overlap with neutral predictions. Nie et al. (2020) use an adversarial human-andmodel-in-the-loop procedure to address these concerns in Adversarial NLI (ANLI). Using a model in the loop makes ANLI inherently adversarial towards the model used, and we instead focus on naturally collected human-in-the-loop augmented data. Kaushik et al. (2020) crowdsource counterfactually-augmented NLI examples that reduce the presence of hypothesis-only bias in SNLI by providing a set of seed examples to crowdworkers and prompting them to minimally edit either the hypothesis or premise to yield a counterfactual label. Khashabi et al. (2020) present this type of data collection as an efficient method to build training sets yielding robust models that generalize better by crowdsourcing counterfactually-augmented BoolQ examples. However, they also find that augmented datasets yield similar to worse performance when the cost of augmenting an example is no cheaper than collecting a new one and the datasets are controlled for size. We differ from Kaushik et al. (2020) by focusing on performance on out-of-domain examples and from Khashabi et al. (2020) by focusing on the task of NLI instead of reading comprehension.",
"cite_spans": [
{
"start": 112,
"end": 133,
"text": "(Poliak et al., 2018;",
"ref_id": "BIBREF13"
},
{
"start": 134,
"end": 158,
"text": "Gururangan et al., 2018;",
"ref_id": "BIBREF4"
},
{
"start": 159,
"end": 174,
"text": "Tsuchiya, 2018)",
"ref_id": "BIBREF14"
},
{
"start": 305,
"end": 311,
"text": "(2018)",
"ref_id": "BIBREF11"
},
{
"start": 425,
"end": 444,
"text": "(Naik et al., 2018;",
"ref_id": null
},
{
"start": 445,
"end": 464,
"text": "McCoy et al., 2019)",
"ref_id": "BIBREF8"
},
{
"start": 688,
"end": 705,
"text": "Nie et al. (2020)",
"ref_id": "BIBREF12"
},
{
"start": 977,
"end": 998,
"text": "Kaushik et al. (2020)",
"ref_id": "BIBREF5"
},
{
"start": 1263,
"end": 1285,
"text": "Khashabi et al. (2020)",
"ref_id": "BIBREF6"
},
{
"start": 1694,
"end": 1715,
"text": "Kaushik et al. (2020)",
"ref_id": "BIBREF5"
},
{
"start": 1778,
"end": 1800,
"text": "Khashabi et al. (2020)",
"ref_id": "BIBREF6"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Related Work",
"sec_num": "2"
},
{
"text": "Gardner et al. (2020) use contrast sets written manually by NLP researchers to evaluate models on various annotated tasks. They show that most datasets require 1-3 minutes per augmented example, taking 17-50 hours to create 1,000 examples. We differ by using crowdsourced counterfactuallyaugmented data and focusing on their use for training instead of evaluation.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Related Work",
"sec_num": "2"
},
{
"text": "We perform two experiments to study the effects of counterfactually-augmented NLI training data. All experiments use RoBERTa trained on SNLI, CNLI, or CNLI seed examples originally sampled from SNLI and compare performances on various tasks. We first compare MNLI performances to evaluate the impact on model generalization to out-ofdomain data. We then use the diagnostic examples from Naik et al. (2018) and the GLUE diagnostic set (Wang et al., 2019a) to study model robustness to challenge examples.",
"cite_spans": [
{
"start": 399,
"end": 405,
"text": "(2018)",
"ref_id": "BIBREF11"
},
{
"start": 434,
"end": 454,
"text": "(Wang et al., 2019a)",
"ref_id": "BIBREF15"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Experimental Setup",
"sec_num": "3"
},
{
"text": "Training Data In SNLI, Bowman et al. (2015) prompt crowdworkers with a scene description premise to collect three hypothesis sentences corresponding to entailment, neutral, and contradiction labels, yielding 570k English premise-hypothesis pairs. Kaushik et al. (2020) collect CNLI examples by prompting crowdworkers to minimally edit seed examples sampled from SNLI to yield counterfactual labels.",
"cite_spans": [
{
"start": 247,
"end": 268,
"text": "Kaushik et al. (2020)",
"ref_id": "BIBREF5"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Experimental Setup",
"sec_num": "3"
},
{
"text": "For our training data, we use a subsampled set of 8.3k examples of SNLI, the CNLI training set of 8.3k examples, and the 1.7k CNLI seed examples sampled from SNLI that is also included in the CNLI training set. We subsample SNLI to control for the fact that CNLI only consists of 8.3k examples. We subsample five sets of 8.3k SNLI examples and report results across these five. and Knowledge. We use these sets to test model robustness to challenge examples. We refer the reader to Naik et al. (2018) and Wang et al. (2019a) for additional details on each diagnostic set.",
"cite_spans": [
{
"start": 494,
"end": 500,
"text": "(2018)",
"ref_id": "BIBREF11"
},
{
"start": 505,
"end": 524,
"text": "Wang et al. (2019a)",
"ref_id": "BIBREF15"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Experimental Setup",
"sec_num": "3"
},
{
"text": "McCoy et al. (2019) provide similar adversarial examples, but we find them too difficult for our models, with performance consistently below 3%, so we do not report performance in detail.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experimental Setup",
"sec_num": "3"
},
{
"text": "Implementation Our code 1 builds on jiant v2 alpha (Wang et al., 2019b) . All experiments use roberta-base. For each round of training, we perform 20 runs and randomly search the hyperparameter space of learning rate {1e-5, 2e-5, 3e-5}, batch size {32, 64}, and random seed. Given the small training set size and stability benefits from longer training found in Mosbach et al. (2020) , we train each run for 20 epochs using early stopping based on the respective validation sets.",
"cite_spans": [
{
"start": 51,
"end": 71,
"text": "(Wang et al., 2019b)",
"ref_id": null
},
{
"start": 362,
"end": 383,
"text": "Mosbach et al. (2020)",
"ref_id": "BIBREF9"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Experimental Setup",
"sec_num": "3"
},
{
"text": "Generalization to MNLI From the median scores in Figure 1 , we see that models trained on CNLI perform no better than models trained on a comparably large sample of unaugmented SNLI examples. This is in line with findings from Khashabi et al. (2020) , where training with their minimally perturbed BoolQ dataset of seed and augmented examples yields similar or worse performance on out-of-domain tasks compared to the original BoolQ training set. Additionally, the improvement of CNLI over the 1.7k seed examples shows that counterfactual examples are somewhat helpful when they are strictly additive, as in Khashabi et al. (2020) . Figure 2 presents performances on the diagnostic sets from Naik et al. (2018) and Wang et al. (2019a) . For the GLUE diagnostic sets, we follow the authors and use R 3 (Gorodkin, 2004) as our evaluation metric. The distributions of classification accuracy again show that CNLI yields similar performance compared to unaugmented datasets of similar size on most of the categories. However, we find that training on CNLI yields worse performance than using either unaugmented SNLI or CNLI seed examples for Negation examples. These challenge examples append the phrase \"and false is not true\" to every hypothesis in the MNLI validation set. This construction introduces the strong negation word \"no\" to target the association between negation words and the contradiction label without changing the truth condition of the sentence. We speculate that the augmented data may have amplified this association already present among the seed examples. Not only does this show that CNLI can yield models that are less robust to certain challenge examples, but it also provides evidence that adding substantial numbers of counterfactual examples to a dataset can hurt robustness.",
"cite_spans": [
{
"start": 227,
"end": 249,
"text": "Khashabi et al. (2020)",
"ref_id": "BIBREF6"
},
{
"start": 608,
"end": 630,
"text": "Khashabi et al. (2020)",
"ref_id": "BIBREF6"
},
{
"start": 692,
"end": 710,
"text": "Naik et al. (2018)",
"ref_id": null
},
{
"start": 715,
"end": 734,
"text": "Wang et al. (2019a)",
"ref_id": "BIBREF15"
},
{
"start": 801,
"end": 817,
"text": "(Gorodkin, 2004)",
"ref_id": "BIBREF3"
}
],
"ref_spans": [
{
"start": 49,
"end": 57,
"text": "Figure 1",
"ref_id": null
},
{
"start": 633,
"end": 641,
"text": "Figure 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Results",
"sec_num": "4"
},
{
"text": "Lexical Diversity Given the minimal edits constraint in CNLI, we study the lexical diversity of the training sets to see the effectiveness of this constraint and whether the general improvement of CNLI over seed examples is a result of greater diversity from a larger training set. Table 1 provides the number of n-grams present in each training set with n varying from one to four. We see that including minimally edited examples to CNLI increases the number of n-grams present, roughly doubling the number of 4-grams, which corresponds to the general improvement over seed examples.",
"cite_spans": [],
"ref_spans": [
{
"start": 282,
"end": 289,
"text": "Table 1",
"ref_id": "TABREF3"
}
],
"eq_spans": [],
"section": "Robustness to Diagnostic Sets",
"sec_num": null
},
{
"text": "We also observe that CNLI contains roughly 70% of 2-, 3-, and 4-grams compared to similarly large unaugmented training sets. This seems natural given the minimal edits constraint when collecting counterfactually-augmented examples and highlights the fact that this type of data augmentation results in less diversity per example.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Robustness to Diagnostic Sets",
"sec_num": null
},
{
"text": "We follow a similar setup to Khashabi et al. (2020) and use English NLI data to test whether counterfactually-augmented training data yields models that generalize better to out-of-domain data and are more robust to challenge examples. We first find that adding counterfactually-augmented data improves generalization, but provides no advantage over adding similar amounts of unaugmented data. Further, we find that the improvement over seed examples corresponds to an increase in n-gram diversity. We also find that including counterfactuallyaugmented data can make models less robust to challenge examples. Assuming that crowdworkers take a similar amount of time to make targeted edits to examples and to write new examples (Bowman et al., 2020) , there is then no obvious value in crowdsourcing augmentations under current protocols for use as training data.",
"cite_spans": [
{
"start": 29,
"end": 51,
"text": "Khashabi et al. (2020)",
"ref_id": "BIBREF6"
},
{
"start": 727,
"end": 748,
"text": "(Bowman et al., 2020)",
"ref_id": "BIBREF1"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "5"
},
{
"text": "Despite these findings, we argue that there is still value in naturally collected counterfactuallyaugmented NLU data. Gardner et al. (2020) show that collecting this type of data can be used as a method to address systematic gaps in testing data. As performances on benchmarks become saturated, we still view this style of augmenting test sets as a viable method to provide longer-lasting benchmarks in addition to standard test set creation.",
"cite_spans": [
{
"start": 118,
"end": 139,
"text": "Gardner et al. (2020)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "5"
},
{
"text": "The success of Gardner et al. (2020) in using expert-designed counterfactual augmentation to target specific phenomena for evaluation suggests that it may be possible to target heuristics in training data with expert guidance during the crowdsourcing process. Further, understanding how to identify heuristics to target and the types of useful augmentations to collect, assuming such a thing is possible, are important directions we leave to future work.",
"cite_spans": [
{
"start": 15,
"end": 36,
"text": "Gardner et al. (2020)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "5"
}
],
"back_matter": [
{
"text": "We thank Clara Vania and Jason Phang for their helpful feedback and Alex Wang for providing the script for n-gram counts that we base our lexical diversity analysis code on. This project has benefited from financial support to SB by Eric and Wendy Schmidt (made by recommendation of the Schmidt Futures program), by Samsung Research (under the project Improving Deep Learning using Latent Structure), by Intuit, Inc., and in-kind support by the NYU High-Performance Computing Center and by NVIDIA Corporation (with the donation of a Titan V GPU). This material is based upon work supported by the National Science Foundation under Grant No. 1922658. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgements",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "A large annotated corpus for learning natural language inference",
"authors": [
{
"first": "R",
"middle": [],
"last": "Samuel",
"suffix": ""
},
{
"first": "Gabor",
"middle": [],
"last": "Bowman",
"suffix": ""
},
{
"first": "Christopher",
"middle": [],
"last": "Angeli",
"suffix": ""
},
{
"first": "Christopher",
"middle": [
"D"
],
"last": "Potts",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Manning",
"suffix": ""
}
],
"year": 2015,
"venue": "Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing",
"volume": "",
"issue": "",
"pages": "632--642",
"other_ids": {
"DOI": [
"10.18653/v1/d15-1075"
]
},
"num": null,
"urls": [],
"raw_text": "Samuel R. Bowman, Gabor Angeli, Christopher Potts, and Christopher D. Manning. 2015. A large an- notated corpus for learning natural language infer- ence. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, EMNLP 2015, Lisbon, Portugal, September 17-21, 2015, pages 632-642. The Association for Compu- tational Linguistics.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "Collecting entailment data for pretraining: New protocols and negative results",
"authors": [
{
"first": "R",
"middle": [],
"last": "Samuel",
"suffix": ""
},
{
"first": "Jennimaria",
"middle": [],
"last": "Bowman",
"suffix": ""
},
{
"first": "Livio Baldini",
"middle": [],
"last": "Palomaki",
"suffix": ""
},
{
"first": "Emily",
"middle": [],
"last": "Soares",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Pitler",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of EMNLP",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Samuel R. Bowman, Jennimaria Palomaki, Livio Bal- dini Soares, and Emily Pitler. 2020. Collecting en- tailment data for pretraining: New protocols and negative results. In Proceedings of EMNLP.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Comparing two k-category assignments by a k-category correlation coefficient",
"authors": [
{
"first": "J",
"middle": [],
"last": "Gorodkin",
"suffix": ""
}
],
"year": 2004,
"venue": "Biology and Chemistry",
"volume": "28",
"issue": "5",
"pages": "367--374",
"other_ids": {
"DOI": [
"10.1016/j.compbiolchem.2004.09.006"
]
},
"num": null,
"urls": [],
"raw_text": "J. Gorodkin. 2004. Comparing two k-category assign- ments by a k-category correlation coefficient. Com- putational Biology and Chemistry, 28(5):367 -374.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "Annotation artifacts in natural language inference data",
"authors": [
{
"first": "Swabha",
"middle": [],
"last": "Suchin Gururangan",
"suffix": ""
},
{
"first": "Omer",
"middle": [],
"last": "Swayamdipta",
"suffix": ""
},
{
"first": "Roy",
"middle": [],
"last": "Levy",
"suffix": ""
},
{
"first": "Samuel",
"middle": [
"R"
],
"last": "Schwartz",
"suffix": ""
},
{
"first": "Noah",
"middle": [
"A"
],
"last": "Bowman",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Smith",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT",
"volume": "2",
"issue": "",
"pages": "107--112",
"other_ids": {
"DOI": [
"10.18653/v1/n18-2017"
]
},
"num": null,
"urls": [],
"raw_text": "Suchin Gururangan, Swabha Swayamdipta, Omer Levy, Roy Schwartz, Samuel R. Bowman, and Noah A. Smith. 2018. Annotation artifacts in natural language inference data. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT, New Orleans, Louisiana, USA, June 1-6, 2018, Volume 2 (Short Papers), pages 107-112. Association for Computa- tional Linguistics.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Learning the difference that makes A difference with counterfactuallyaugmented data",
"authors": [
{
"first": "Divyansh",
"middle": [],
"last": "Kaushik",
"suffix": ""
},
{
"first": "Eduard",
"middle": [
"H"
],
"last": "Hovy",
"suffix": ""
},
{
"first": "Zachary",
"middle": [
"Chase"
],
"last": "Lipton",
"suffix": ""
}
],
"year": 2020,
"venue": "8th International Conference on Learning Representations",
"volume": "2020",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Divyansh Kaushik, Eduard H. Hovy, and Zachary Chase Lipton. 2020. Learning the differ- ence that makes A difference with counterfactually- augmented data. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "More bang for your buck: Natural perturbation for robust question answering",
"authors": [
{
"first": "Daniel",
"middle": [],
"last": "Khashabi",
"suffix": ""
},
{
"first": "Tushar",
"middle": [],
"last": "Khot",
"suffix": ""
},
{
"first": "Ashish",
"middle": [],
"last": "Sabharwal",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of EMNLP",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Daniel Khashabi, Tushar Khot, and Ashish Sabharwal. 2020. More bang for your buck: Natural perturba- tion for robust question answering. In Proceedings of EMNLP.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Right for the wrong reasons: Diagnosing syntactic heuristics in natural language inference",
"authors": [
{
"first": "Tom",
"middle": [],
"last": "Mccoy",
"suffix": ""
},
{
"first": "Ellie",
"middle": [],
"last": "Pavlick",
"suffix": ""
},
{
"first": "Tal",
"middle": [],
"last": "Linzen",
"suffix": ""
}
],
"year": 2019,
"venue": "Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019",
"volume": "1",
"issue": "",
"pages": "3428--3448",
"other_ids": {
"DOI": [
"10.18653/v1/p19-1334"
]
},
"num": null,
"urls": [],
"raw_text": "Tom McCoy, Ellie Pavlick, and Tal Linzen. 2019. Right for the wrong reasons: Diagnosing syntactic heuristics in natural language inference. In Pro- ceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28-August 2, 2019, Volume 1: Long Pa- pers, pages 3428-3448. Association for Computa- tional Linguistics.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "On the stability of fine-tuning BERT: misconceptions, explanations, and strong baselines. CoRR, abs",
"authors": [
{
"first": "Marius",
"middle": [],
"last": "Mosbach",
"suffix": ""
},
{
"first": "Maksym",
"middle": [],
"last": "Andriushchenko",
"suffix": ""
},
{
"first": "Dietrich",
"middle": [],
"last": "Klakow",
"suffix": ""
}
],
"year": 2006,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Marius Mosbach, Maksym Andriushchenko, and Diet- rich Klakow. 2020. On the stability of fine-tuning BERT: misconceptions, explanations, and strong baselines. CoRR, abs/2006.04884.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "Stress test evaluation for natural language inference",
"authors": [],
"year": 2018,
"venue": "Proceedings of the 27th International Conference on Computational Linguistics",
"volume": "",
"issue": "",
"pages": "2340--2353",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Stress test evaluation for natural language in- ference. In Proceedings of the 27th International Conference on Computational Linguistics, COLING 2018, Santa Fe, New Mexico, USA, August 20-26, 2018, pages 2340-2353. Association for Computa- tional Linguistics.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Adversarial NLI: A new benchmark for natural language understanding",
"authors": [
{
"first": "Yixin",
"middle": [],
"last": "Nie",
"suffix": ""
},
{
"first": "Adina",
"middle": [],
"last": "Williams",
"suffix": ""
},
{
"first": "Emily",
"middle": [],
"last": "Dinan",
"suffix": ""
},
{
"first": "Mohit",
"middle": [],
"last": "Bansal",
"suffix": ""
},
{
"first": "Jason",
"middle": [],
"last": "Weston",
"suffix": ""
},
{
"first": "Douwe",
"middle": [],
"last": "Kiela",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
"volume": "2020",
"issue": "",
"pages": "4885--4901",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yixin Nie, Adina Williams, Emily Dinan, Mohit Bansal, Jason Weston, and Douwe Kiela. 2020. Ad- versarial NLI: A new benchmark for natural lan- guage understanding. In Proceedings of the 58th An- nual Meeting of the Association for Computational Linguistics, ACL 2020, Online, July 5-10, 2020, pages 4885-4901. Association for Computational Linguistics.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "Hypothesis only baselines in natural language inference",
"authors": [
{
"first": "Adam",
"middle": [],
"last": "Poliak",
"suffix": ""
},
{
"first": "Jason",
"middle": [],
"last": "Naradowsky",
"suffix": ""
},
{
"first": "Aparajita",
"middle": [],
"last": "Haldar",
"suffix": ""
},
{
"first": "Rachel",
"middle": [],
"last": "Rudinger",
"suffix": ""
},
{
"first": "Benjamin",
"middle": [],
"last": "Van Durme",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics, *SEM@NAACL-HLT 2018",
"volume": "",
"issue": "",
"pages": "180--191",
"other_ids": {
"DOI": [
"10.18653/v1/s18-2023"
]
},
"num": null,
"urls": [],
"raw_text": "Adam Poliak, Jason Naradowsky, Aparajita Haldar, Rachel Rudinger, and Benjamin Van Durme. 2018. Hypothesis only baselines in natural language in- ference. In Proceedings of the Seventh Joint Con- ference on Lexical and Computational Semantics, *SEM@NAACL-HLT 2018, New Orleans, Louisiana, USA, June 5-6, 2018, pages 180-191. Association for Computational Linguistics.",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "Performance impact caused by hidden bias of training data for recognizing textual entailment",
"authors": [
{
"first": "Masatoshi",
"middle": [],
"last": "Tsuchiya",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of the Eleventh International Conference on Language Resources and Evaluation, LREC 2018",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Masatoshi Tsuchiya. 2018. Performance impact caused by hidden bias of training data for rec- ognizing textual entailment. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation, LREC 2018, Miyazaki, Japan, May 7-12, 2018. European Language Re- sources Association (ELRA).",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "GLUE: A multi-task benchmark and analysis platform for natural language understanding",
"authors": [
{
"first": "Alex",
"middle": [],
"last": "Wang",
"suffix": ""
},
{
"first": "Amanpreet",
"middle": [],
"last": "Singh",
"suffix": ""
},
{
"first": "Julian",
"middle": [],
"last": "Michael",
"suffix": ""
},
{
"first": "Felix",
"middle": [],
"last": "Hill",
"suffix": ""
},
{
"first": "Omer",
"middle": [],
"last": "Levy",
"suffix": ""
},
{
"first": "Samuel",
"middle": [
"R"
],
"last": "Bowman",
"suffix": ""
}
],
"year": 2019,
"venue": "7th International Conference on Learning Representations",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Alex Wang, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel R. Bowman. 2019a. GLUE: A multi-task benchmark and analysis plat- form for natural language understanding. In 7th International Conference on Learning Representa- tions, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019. OpenReview.net.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "2019b. jiant 1.3: A software toolkit for research on general-purpose text understanding models",
"authors": [
{
"first": "Alex",
"middle": [],
"last": "Wang",
"suffix": ""
},
{
"first": "Ian",
"middle": [
"F"
],
"last": "Tenney",
"suffix": ""
},
{
"first": "Yada",
"middle": [],
"last": "Pruksachatkun",
"suffix": ""
},
{
"first": "Phil",
"middle": [],
"last": "Yeres",
"suffix": ""
},
{
"first": "Jason",
"middle": [],
"last": "Phang",
"suffix": ""
},
{
"first": "Haokun",
"middle": [],
"last": "Liu",
"suffix": ""
},
{
"first": ",",
"middle": [],
"last": "Phu Mon Htut",
"suffix": ""
},
{
"first": "Katherin",
"middle": [],
"last": "Yu",
"suffix": ""
},
{
"first": "Jan",
"middle": [],
"last": "Hula",
"suffix": ""
},
{
"first": "Patrick",
"middle": [],
"last": "Xia",
"suffix": ""
},
{
"first": "Raghu",
"middle": [],
"last": "Pappagari",
"suffix": ""
},
{
"first": "R",
"middle": [
"Thomas"
],
"last": "Shuning Jin",
"suffix": ""
},
{
"first": "Roma",
"middle": [],
"last": "Mccoy",
"suffix": ""
},
{
"first": "Yinghui",
"middle": [],
"last": "Patel",
"suffix": ""
},
{
"first": "Edouard",
"middle": [],
"last": "Huang",
"suffix": ""
},
{
"first": "Najoung",
"middle": [],
"last": "Grave",
"suffix": ""
},
{
"first": "Thibault",
"middle": [],
"last": "Kim",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "F\u00e9vry",
"suffix": ""
}
],
"year": null,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Alex Wang, Ian F. Tenney, Yada Pruksachatkun, Phil Yeres, Jason Phang, Haokun Liu, Phu Mon Htut, , Katherin Yu, Jan Hula, Patrick Xia, Raghu Pap- pagari, Shuning Jin, R. Thomas McCoy, Roma Pa- tel, Yinghui Huang, Edouard Grave, Najoung Kim, Thibault F\u00e9vry, Berlin Chen, Nikita Nangia, Anhad Mohananey, Katharina Kann, Shikha Bordia, Nico- las Patry, David Benton, Ellie Pavlick, and Samuel R. Bowman. 2019b. jiant 1.3: A software toolkit for research on general-purpose text understanding models. http://jiant.info/.",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "A broad-coverage challenge corpus for sentence understanding through inference",
"authors": [
{
"first": "Adina",
"middle": [],
"last": "Williams",
"suffix": ""
},
{
"first": "Nikita",
"middle": [],
"last": "Nangia",
"suffix": ""
},
{
"first": "Samuel",
"middle": [
"R"
],
"last": "",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2018",
"volume": "1",
"issue": "",
"pages": "1112--1122",
"other_ids": {
"DOI": [
"10.18653/v1/n18-1101"
]
},
"num": null,
"urls": [],
"raw_text": "Adina Williams, Nikita Nangia, and Samuel R. Bow- man. 2018. A broad-coverage challenge corpus for sentence understanding through inference. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computa- tional Linguistics: Human Language Technologies, NAACL-HLT 2018, New Orleans, Louisiana, USA, June 1-6, 2018, Volume 1 (Long Papers), pages 1112-1122. Association for Computational Linguis- tics.",
"links": null
}
},
"ref_entries": {
"TABREF3": {
"num": null,
"html": null,
"type_str": "table",
"content": "<table/>",
"text": "Number of unique n-gram types observed in each training set."
}
}
}
} |