On Adversarial Removal of Hypothesis-only Bias in Natural Language Inference
Open accessThe DOI matches its Crossref record; 6 of the 6 references it cites were found in a registry or an index.
Abstract
Popular Natural Language Inference (NLI) datasets have been shown to be tainted by hypothesis-only biases.Adversarial learning may help models ignore sensitive biases and spurious correlations in data.We evaluate whether adversarial learning can be used in NLI to encourage models to learn representations free of hypothesis-only biases.Our analyses indicate that the representations learned via adversarial learning may be less biased, with only small drops in NLI accuracy.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
A large annotated corpus for learning natural language inference
Bowman, S. R., Angeli, G., Potts, C., & Manning, C. D. (2015) · Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d15-1075 · Cited by 3,377 · cited 2×
“(2018) found that the Stanford Natural Language Inference dataset (SNLI; Bowman et al., 2015) contained the most (or worst) hypothesisonly biases—their hypothesis-only model outperformed the majority baseline by roughly 100% (going from roughly 34% to 69%).”
BackgroundMethodSupervised Learning of Universal Sentence Representations from Natural Language Inference Data
Conneau, A., Kiela, D., Schwenk, H., Barrault, L., & Bordes, A. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1070 · Cited by 2,049 · cited 10×
“4 , λ Enc = 1 ) predicts contra-diction much less frequently than InferSent on examples with these words.”
MethodInfluentialMitigating Unwanted Biases with Adversarial Learning
Zhang, B. H., Lemoine, B., & Mitchell, M. (2018) · Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3278721.3278779 · Cited by 1,387 · cited 2×
“…approach, referred to as AdvCls, follows the common adversarial training method (Good-fellow et al., 2015; Ganin & Lempitsky, 2015; Xie et al., 2017; Zhang et al., 2018) by adding an additional adversarial classifier c Hypoth to our model. c Hypoth maps the hypothesis representation h to an output…”
MethodAnnotation Artifacts in Natural Language Inference Data
Gururangan, S., Swayamdipta, S., Levy, O., Schwartz, R., Bowman, S., & Smith, N. A. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers) · doi:10.18653/v1/n18-2017 · Cited by 907
What you can cram into a single vector: Probing sentence embeddings for linguistic properties
Conneau, A., Kruszewski, G., Lample, G., Barrault, L., & Baroni, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.01070 · Cited by 263
Cited by
WinoGrande: An Adversarial Winograd Schema Challenge at Scale
Sakaguchi, K., Le Bras, R., Bhagavatula, C., & Choi, Y. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6399 · Cited by 579
Don’t Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases
Clark, C., Yatskar, M., & Zettlemoyer, L. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1418 · Cited by 381
“…(Lewis and Fan, 2019), carefully designed model architecture (Agrawal et al., 2018; Zhang et al., 2016), or adversarial removal of class-indicative features from model’s internal representations (Ramakrish-nan et al., 2018; Zhang et al., 2018a; Belinkov et al., 2019; Grand and Belinkov, 2019).”
BackgroundWinoGrande: an adversarial winograd schema challenge at scale
Sakaguchi, K., Bras, R. L., Bhagavatula, C., & Choi, Y. (2021) · Communications of the ACM · doi:10.1145/3474381 · Cited by 356
“Following a rigorous process, a question is deemed valid if (1) the majority of the three workers chooses the correct answer option, (2) they agree that the two answer options are unambiguous (one option is clearly more plausible than the other), and (3) the question cannot be answered simply by word association in wh…”
BackgroundMethodResultInfluentialProbing the Probing Paradigm: Does Probing Accuracy Entail Task Relevance?
Ravichander, A., Belinkov, Y., & Hovy, E. (2021) · Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume · doi:10.18653/v1/2021.eacl-main.295 · Cited by 132
Self-citationDon’t Take the Premise for Granted: Mitigating Artifacts in Natural Language Inference
Belinkov, Y., Poliak, A., Shieber, S., Van Durme, B., & Rush, A. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1084 · Cited by 73
Self-citationStretching Sentence-pair NLI Models to Reason over Long Documents and Clusters
Schuster, T., Chen, S., Buthpitiya, S., Fabrikant, A., & Metzler, D. (2022) · Findings of the Association for Computational Linguistics: EMNLP 2022 · doi:10.18653/v1/2022.findings-emnlp.28 · Cited by 17
Metadata
- Authors
- Belinkov, Y., Poliak, A., Shieber, S., Van Durme, B., & Rush, A.
- Venue
- Proceedings of the Eighth Joint Conference on Lexical and Computational Semantics (*SEM 2019)
- Published
- 2019
- Pages
- 256-262
- OpenAlex
- W2955503152
- Cited by
- 56
- References
- 6
- FWCI
- 4.01
- Type
- conference paper
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.