Dominican Republic

Understanding robustness and sensitivity of BERT models predicting Alzheimer’s disease from text is important for both developing better classification models and for understanding their capabilities and limitations. In this paper, we analyze how a controlled amount of desired and undesired text alterations impacts performance of BERT.We show that BERT is robust to natural linguistic variations in text. On the other hand, we show that BERT is not sensitive to removing clinically important information from text.

EMNLP 2021

Robustness and Sensitivity of BERT Models Predicting Alzheimer&#39;s Disease from Text

EMNLP 2021 is planned to be a hybrid event in Punta Cana, Dominican Republic, with both on-site and fully virtual participation possible. The experience for on-site participants would closely approximate a normal pre-COVID *ACL conference, with 5-6 thematically organized parallel sessions and live Q/A and interactive discussion immediately after the talks. Presentations by virtual participants will be equitably interleaved with those of on-site participants, projected on the auditorium screens as if on-site, and also followed immediately by live Q/A and interactive discussion at a time during reasonable waking hours for the virtual presenter. For all participants, on-site and virtual, who are unable to attend a session due to either time-zone issues or because they are participating in another session live, talk recordings and slides will be available online at a minimum after the live presentation (and in many cases before as well), and questions may be submitted in advance on session-specific discussion boards and answered live in session with the usual visual aids if desired.

<iframe style="width:700px;height:400px" src="https://online.fliphtml5.com/ebtyf/ceby/"  seamless="seamless" scrolling="no" frameborder="0" allowtransparency="true" allowfullscreen="true" ></iframe>

Please Note: The EMNLP registration system is not currently connected to the underline site as we are still in the process of building out EMNLP 2021. You will receive access instructions from underline the week of November 1st. 

Access is given only to EMNLP upon registration, if you have not registered please do so [here](https://2021.emnlp.org/registration).

Registered attendees will receive access the week of November 1st.

EMNLP 2021 is planned to be a hybrid event in Punta Cana, Dominican Republic, with both on-site and fully virtual participation possible.

Robustness and Sensitivity of BERT Models Predicting Alzheimer's Disease from Text

workshop paper

Dialogue-based relation extraction (DiaRE) aims to detect the structural information from unstructured utterances in dialogues. Existing relation extraction models may be unsatisfactory under such a conversational setting, due to the entangled logic and information sparsity issues in utterances involving multiple speakers. To this end, we introduce SOLS, a novel model which can explicitly induce speaker-oriented latent structures for better DiaRE. Specifically, we learn latent structures to capture the relationships among tokens beyond the utterance boundaries, alleviating the entangled logic issue. During the learning process, our speaker-specific regularization method progressively highlights speaker-related key clues and erases the irrelevant ones, alleviating the information sparsity issue. Experiments on three public datasets demonstrate the effectiveness of our proposed approach.

Speaker-Oriented Latent Structures for Dialogue-Based Relation Extraction

Span extraction, aiming to extract text spans (such as words or phrases) from plain texts, is a fundamental process in Information Extraction. Recent works introduce the label knowledge to enhance the text representation by formalizing the span extraction task into a question answering problem (QA Formalization), which achieves state-of-the-art performance. However, QA Formalization does not fully exploit the label knowledge and suffers from low efficiency in training/inference. To address those problems, we introduce a new paradigm to integrate label knowledge and further propose a novel model to explicitly and efficiently integrate label knowledge into text representations. Specifically, it encodes texts and label annotations independently and then integrates label knowledge into text representation with an elaborate-designed semantics fusion module. We conduct extensive experiments on three typical span extraction tasks: flat NER, nested NER, and event detection. The empirical results show that 1) our method achieves state-of-the-art performance on four benchmarks, and 2) reduces training time and inference time by 76% and 77% on average, respectively, compared with the QA Formalization paradigm.

Enhanced Language Representation with Label Knowledge for Span Extraction

Low-resource Relation Extraction (LRE) aims to extract relation facts from limited labeled corpora when human annotation is scarce. Existing works either utilize self-training scheme to generate pseudo labels that will cause the gradual drift problem, or leverage metalearning scheme which does not solicit feedback explicitly. To alleviate selection bias due to the lack of feedback loops in existing LRE learning paradigms, we developed a Gradient Imitation Reinforcement Learning method to encourage pseudo label data to imitate the gradient descent direction on labeled data and bootstrap its optimization capability through trial and error. We also propose a framework called GradLRE, which handles two major scenarios in low-resource relation extraction. Besides the scenario where unlabeled data is sufficient, GradLRE handles the situation where no unlabeled data is available, by exploiting a contextualized augmentation method to generate data. Experimental results on two public datasets demonstrate the effectiveness of GradLRE on low resource relation extraction when comparing with baselines.

Gradient Imitation Reinforcement Learning for Low Resource Relation Extraction

Taxonomies are symbolic representations of hierarchical relationships between terms or entities. While taxonomies are useful in broad applications, manually updating or maintaining them is labor-intensive and difficult to scale in practice. Conventional supervised methods for this enrichment task fail to find optimal parents of new terms in low-resource settings where only small taxonomies are available because of overfitting to hierarchical relationships in the taxonomies. To tackle the problem of low-resource taxonomy enrichment, we propose Musubu, an efficient framework for taxonomy enrichment in low-resource settings with pretrained language models (LMs) as knowledge bases to compensate for the shortage of information. Musubu leverages an LM-based classifier to determine whether or not inputted term pairs have hierarchical relationships. Musubu also utilizes Hearst patterns to generate queries to leverage implicit knowledge from the LM efficiently for more accurate prediction. We empirically demonstrate the effectiveness of our method in extensive experiments on taxonomies from both a SemEval task and real-world retailer datasets.

Low-resource Taxonomy Enrichment with Pretrained Language Models

The clustering-based unsupervised relation discovery method has gradually become one of the important methods of open relation extraction (OpenRE). However, high-dimensional vectors can encode complex linguistic information which leads to the problem that the derived clusters cannot explicitly align with the relational semantic classes. In this work, we propose a relation-oriented clustering model and use it to identify the novel relations in the unlabeled data. Specifically, to enable the model to learn to cluster relational data, our method leverages the readily available labeled data of pre-defined relations to learn a relation-oriented representation. We minimize distance between the instance with same relation by gathering the instances towards their corresponding relation centroids to form a cluster structure, so that the learned representation is cluster-friendly. To reduce the clustering bias on predefined classes, we optimize the model by minimizing a joint objective on both labeled and unlabeled data. Experimental results show that our method reduces the error rate by 29.2% and 15.7%, on two datasets respectively, compared with current SOTA methods.

A Relation-Oriented Clustering Method for Open Relation Extraction

Most of the existing studies of language use in social media content have focused on the surface-level linguistic features and the semantic level aspects of the comments. The writer‘s strategies of constructing and connecting text segments have not been widely explored. Contributing to this analysis direction for social media studies, we build a comparable neural RST parsing system that analyzes discourse relations in persuasive and non-persuasive comments. Our work demonstrates the potential of analyzing discourse structures of online comments with our system and the implications of these structures for understanding online communications.

Neural-based RST Parsing And Analysis In Persuasive Discourse

Optical character recognition (OCR) from newspaper page images is susceptible to noise due to degradation of old documents and variation in typesetting. In this report, we present a novel approach to OCR post-correction. We cast error correction as a translation task, and fine-tune BART, a transformer-based sequence-to-sequence language model pretrained to denoise corrupted text. We are the first to use sentence-level transformer models for OCR post-correction, and our best model achieves a 29.4% improvement in character accuracy over the original noisy OCR text. Our results demonstrate the utility of pretrained language models for dealing with noisy text.

BART for Post-Correction of OCR Newspaper Text

Stance detection (SD) entails classifying the sentiment of a text towards a given target, and is a relevant sub-task for opinion mining and social media analysis. Recent works have explored knowledge infusion - supplementing the linguistic competence and latent knowledge of large pre-trained language models with structured knowledge graphs (KGs), yet few works have applied such methods to the SD task. In this work, we first perform stance-relevant knowledge probing on Transformers-based pre-trained models in a zero-shot setting, showing these models' latent real-world knowledge about SD targets and their sensitivity to context. We then propose novel knowledge-enriched stance detection models. We evaluate them on two Twitter stance datasets, achieving state-of-the-art performance on both.

Integrating Transformers and Knowledge Graphs for Twitter Stance Detection

Online social media platforms increasinglyrely on Natural Language Processing (NLP)techniques to detect abusive content at scalein order to mitigate the harms it causes totheir users. However, these techniques suf-fer from various sampling and association bi-ases present in training data, often resultingin sub-par performance on content relevantto marginalized groups, potentially furtheringdisproportionate harms towards them. Studieson such biases so far have focused on only ahandful of axes of disparities and subgroupsthat have annotations/lexicons available. Con-sequently, biases concerning non-Western con-texts are largely ignored in the literature. Inthis paper, we introduce a weakly supervisedmethod to robustly detect lexical biases inbroader geocultural contexts. Through a casestudy on a publicly available toxicity detec-tion model, we demonstrate that our methodidentifies salient groups of cross-geographicerrors, and, in a follow up, demonstrate thatthese groupings reflect human judgments of of-fensive and inoffensive language in those geo-graphic contexts. We also conduct analysis ofa model trained on a dataset with ground truthlabels to better understand these biases, andpresent preliminary mitigation experiments.

Detecting and Evaluating Cross-Geographic Biases in Toxicity Modelingon Social Media Text

On Wikipedia, an online crowdsourced encyclopedia, volunteers enforce the encyclopedia’s editorial policies. Wikipedia’s policy on maintaining a neutral point of view has inspired recent research on bias detection, including “weasel words” and “hedges”. Yet to date, little work has been done on identifying “puffery,” phrases that are overly positive without a verifiable source. We demonstrate that collecting training data for this task requires some care, and construct a dataset by com-
bining Wikipedia editorial annotations and information retrieval techniques. We compare several approaches to predicting puffery, and achieve 0.963 f1 score by incorporating citation features into a RoBERTa model. Finally, we demonstrate how to integrate our model with Wikipedia’s public infrastructure to give back to the Wikipedia editor community.

Content not yet available

Robustness and Sensitivity of BERT Models Predicting Alzheimer's Disease from Text

Next from EMNLP 2021

Speaker-Oriented Latent Structures for Dialogue-Based Relation Extraction

Similar lecture

{HOTTER}: {H}ierarchical Optimal Topic Transport with Explanatory Context Representations