The KnowRef Coreference Corpus: Removing Gender and Number Cues for Difficult Pronominal Anaphora Resolution

By Ali Emami, Paul Trichelair, Adam Trischler, Kaheer Suleman, Hannes Schulz, Jackie Chi Kit Cheung. 2018

58 citations recorded by the source corpus.

We introduce a new benchmark for coreference resolution and NLI, KnowRef, that targets common-sense understanding and world knowledge. Previous coreference resolution tasks can largely be solved by exploiting the number and gender of the antecedents, or have been handcrafted and do not reflect the diversity of naturally occurring text. We present a corpus of over 8,000 annotated text passages with ambiguous pronominal anaphora. These instances are both challenging and realistic. We show that various coreference systems, whether rule-based, feature-rich, or neural, perform significantly worse on the task than humans, who display high inter-annotator agreement. To explain this performance gap, we show empirically that state-of-the art models often fail to capture context, instead relying on the gender or number of candidate antecedents to make a decision. We then use problem-specific insights to propose a data-augmentation trick called antecedent switching to alleviate this tendency in models. Finally, we show that antecedent switching yields promising results on other tasks as well: we use it to achieve state-of-the-art results on the GAP coreference task.

Subfields: Artificial Intelligence, Natural Language Processing, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Evaluation & Benchmarks

Citations in the corpus

Queens Are Powerful Too: Mitigating Gender Bias in Dialogue Generation (2019, background): “2019), uncovering many gender disparities and harmful biases in algorithms and text (Cao and Daumé III 2020; Chang et al. 2019; Chang and McKeown 2019; Costa-jussà 2019; Du et al. 2019; Emami et al. 2019; Garimella et al. 2019; Gaut et al. 2020; Habash et al. 2019; Hashempour 2019; Hoyle et al. 2019”
Multi-Dimensional Gender Bias Classification (2020, background): “, 2019)—and affect a wide range of downstream tasks including coreference resolution (Zhao et al., 2018a; Cao and Daumé, 2019; Emami et al., 2019), part-of-speech and dependency parsing (Garimella et al.”
A Survey on Gender Bias in Natural Language Processing (2021, background): “Since the model observes the same scenario in the doubled (for binary gender) sentences, it can learn to abstract away from the entities to the context [Emami et al. 2019].”
Contrastive Self-Supervised Learning for Commonsense Reasoning (2020, background): “(Davis et al., 2016 BERT-ft (Kocijan et al., 2019) 78.3 % PDP-60 (unsupervised) Unsupervised Sem.”
WikiCREM: A Large Unsupervised Corpus for Coreference Resolution (2019, methodology): “We did not evaluate on the KNOWREF dataset (Emami et al., 2019), since it was not yet publicly available at the time of writing.”
Pre-training of Mention Representations in Coreference Models (2020, background): “Emami et al. (2019) use names and pronouns gender information to generate links between a pronoun and a name.”
Entity-Switched Datasets: An Approach to Auditing the In-Domain Robustness of Named Entity Recognition Models (2020, background): “…et al., 2019) to study the impact of person names on toxicity in online comments by substituting in names of controversial personalities and in (Emami et al., 2019) to make coreference resolution robust to gender and number cues by making both antecedent and candidate of the same type.”