Skip to search formSkip to main contentSkip to account menu

Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection

This work presents Iterative Null-space Projection (INLP), a novel method for removing information from neural representations based on repeated training of linear classifiers that predict a certain property the authors aim to remove, followed by projection of the representations on their null-space.
View on ACL (opens in a new tab)

oLMpics-On What Language Model Pre-training Captures

This work proposes eight reasoning tasks, which conceptually require operations such as comparison, conjunction, and composition, and findings can help future work on designing new datasets, models, and objective functions for pre-training.
View on MIT Press (opens in a new tab)

Adversarial Removal of Demographic Attributes from Text Data

It is shown that demographic information of authors is encoded in—and can be recovered from—the intermediate representations learned by text-based neural classifiers, and the implication is that decisions of classifiers trained on textual data are not agnostic to—and likely condition on—demographic attributes.
View on ACL (opens in a new tab)

Amnesic Probing: Behavioral Explanation with Amnesic Counterfactuals

The inability to infer behavioral conclusions from probing results is pointed out, and an alternative method that focuses on how the information is being used is offered, rather than on what information is encoded is offered.
View on MIT Press (opens in a new tab)

Measuring and Improving Consistency in Pretrained Language Models

The creation of PARAREL, a high-quality resource of cloze-style query English paraphrases, and analysis of the representational spaces of PLMs suggest that they have a poor structure and are currently not suitable for representing knowledge in a robust way.
View on MIT Press (opens in a new tab)

Evaluating Models’ Local Decision Boundaries via Contrast Sets

A more rigorous annotation paradigm for NLP that helps to close systematic gaps in the test data, and recommends that the dataset authors manually perturb the test instances in small but meaningful ways that (typically) change the gold label, creating contrast sets.
View on ACL (opens in a new tab)

Do Language Embeddings capture Scales?

This work identifies contextual information in pre-training and numeracy as two key factors affecting their performance, and shows that a simple method of canonicalizing numbers can have a significant effect on the results.
View on ACL (opens in a new tab)

Evaluating NLP Models via Contrast Sets

A new annotation paradigm for NLP is proposed that helps to close systematic gaps in the test data, and it is recommended that after a dataset is constructed, the dataset authors manually perturb the test instances in small but meaningful ways that change the gold label, creating contrast sets.
View PDF on arXiv (opens in a new tab)

Adversarial Removal of Demographic Attributes Revisited

It is shown that a diagnostic classifier trained on the biased baseline neural network also does not generalize to new samples, indicating that it relies on correlations specific to their particular data sample.
View on ACL (opens in a new tab)

Contrastive Explanations for Model Interpretability

The ability of label-contrastive explanations to provide fine-grained interpretability of model decisions is demonstrated, via both high-level abstract concept attribution and low-level input token/span attribution for two NLP classification benchmarks.
View on ACL (opens in a new tab)
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)