Spain

As the name implies, contextualized representations of language are typically motivated by their ability to encode context. Which aspects of context are captured by such representations? We introduce an approach to address this question using Representational Similarity Analysis (RSA). As case studies, we investigate the degree to which a verb embedding encodes the verb’s subject, a pronoun embedding encodes the pronoun’s antecedent, and a full-sentence representation encodes the sentence’s head word (as determined by a dependency parse). In all cases, we show that BERT’s contextualized embeddings reflect the linguistic dependency being studied, and that BERT encodes these dependencies to a greater degree than it encodes less linguistically-salient controls. These results demonstrate the ability of our approach to adjudicate between hypotheses about which aspects of context are encoded in representations of language.

COLING 2020

Picking BERT’s Brain: Probing for Linguistic Dependencies in Contextualized Embeddings Using Representational Similarity Analysis

stages 

COLING, the International Conference on Computational Linguistics, is one of the premier conferences for natural language processing and computational linguistics. Often grouped within the field of artificial intelligence, but actually pre-dating the development of artificial intelligence, advances in computational linguistics and natural language processing are now some of the major drivers behind the use of artificial intelligence for commercial and social applications – for example, on-line search, machine translation and with voice-assisted conversational devices.

First established in 1965, the biennial COLING conference is held in diverse parts of the globe and attracts participants from both top-ranked research centers and emerging countries. Today, the most important developments in our field are taking place not only in universities and academic research institutes, but also in industrial research departments and in technological startups. COLING conferences provide opportunities for all these communities to showcase their exciting developments.

COLING, the International Conference on Computational Linguistics, is one of the premier conferences for natural language processing and computational linguistics. Often grouped within the field of artificial intelligence, but actually pre-dating the development of artificial intelligence, advances in computational linguistics and natural language processing are now some of the major drivers behind the use of artificial intelligence for commercial and social applications – for example, on-line search, machine translation and with voice-assisted conversational devices.

poster

Sign Language Translation (SLT) first uses a Sign Language Recognition (SLR) system to extract sign language glosses from videos. Then, a translation system generates spoken language translations from the sign language glosses. Previous work improves the first system but little work has focused on the second. This paper focuses on the translation system and introduces the STMC-Transformer which improves on the current state-of-the-art by over 5 and 7 BLEU respectively on gloss-to-text and video-to-text translation of the PHOENIX-Weather 2014T dataset. On the ASLG-PC12 corpus, we report an increase of over 16 BLEU.
We also demonstrate the problem in current methods that rely on gloss supervision. The video-to-text translation of our STMC-Transformer outperforms translation of ground truth (GT) glosses. This contradicts previous claims that GT gloss translation acts as an upper bound for SLT performance and reveals that glosses are an inefficient representation of sign language. For future SLT research, we suggest an end-to-end training of the recognition and translation models or using a different sign language annotation scheme.

Better Sign Language Translation with STMC-Transformer

In this work we present the experiments which lead to the creation of our BERT and ELECTRA based German language models, GBERT and GELECTRA. By varying the input training data, model size, and the presence of Whole Word Masking (WWM) we were able to attain SoTA performance across a set of document classification and named entity recognition (NER) tasks for both models of base and large size. We adopt an evaluation driven approach in training these models and our results indicate that both adding more data and utilizing WWM improve model performance. By benchmarking against existing German models, we show that these models are the best German models to date. Our trained models will be made publicly available to the research community.

German’s Next Language Model

Emotion-cause pair extraction (ECPE), which aims at simultaneously extracting emotion-cause pairs that express emotions and their corresponding causes in a document, plays a vital role in understanding natural languages. Considering that a cause usually appears around its corresponding emotion, we construct a pair graph and a Pair Graph Convolutional Network (PairGCN) to model dependency relations among local neighborhood candidate pairs. Moreover, in our proposed graph, there are three types of dependency relations and each type of dependency relations has its own way to propagate contextual information. Experiments on a benchmark Chinese emotion-cause pair extraction corpus demonstrate the effectiveness of the proposed model.

End-to-End Emotion-Cause Pair Extraction with Graph Convolutional Network

This paper describes the system designed by the Baidu PGL Team which achieved the first place in the TextGraphs 2020 Shared Task. The task focuses on generating explanations for elementary science questions. Given a question and its corresponding correct answer, we are asked to select the facts that can explain why the answer is correct for that question and answering (QA) from a large knowledge base. To address this problem, we use a pre-trained language model to recall the top-K relevant explanations for each question. Then, we adopt a re-ranking approach based on a pre-trained language model to rank the candidate explanations. To further improve the rankings, we also develop an architecture consisting both powerful pre-trained transformers and GNNs to tackle the multi-hop inference problem. The official evaluation shows that, our system can outperform the second best system by 1.91 points. 

Explanation Regeneration using Language and Graph Learning Methods

The experimental landscape in natural language processing for social media is too fragmented. Each year, new shared tasks and datasets are proposed, ranging from classics like sentiment analysis to irony detection or emoji prediction. Therefore, it is unclear what the current state of the art is, as there is no standardized evaluation protocol, neither a strong set of baselines trained on such domain- specific data. In this paper, we propose a new evaluation framework (TWEETEVAL) consist- ing of seven heterogeneous Twitter-specific classification tasks. We also provide a strong set of baselines as starting point, and com- pare different language modeling pre-training strategies. Our initial experiments show the effectiveness of starting off with existing pre- trained generic language models, and continue training them on Twitter corpora.

TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classification

It is an open question to what extent perceptions of literary quality are derived from text-intrinsic versus social factors. While supervised models can predict literary quality ratings from textual factors quite successfully, as shown in the Riddle of Literary Quality project (Koolen et al., 2020), this does not prove that social factors are not important, nor can we assume that readers make judgments on literary quality in the same way and based on the same information as machine learning models. We report the results of a pilot study to gauge the effect of textual features on literary ratings of Dutch-language novels by participants in a controlled experiment with 48 participants. In an exploratory analysis, we compare the ratings to those from the large reader survey of the Riddle in which social factors were not excluded, and to machine learning predictions of those literary ratings. We find moderate to strong correlations of questionnaire ratings with the survey ratings, but the predictions are closer to the survey ratings. Code and data: https://github.com/andreasvc/litquest

Results of a Single Blind Literary Taste Test

This paper presents AutoGOAL, a system for automatic machine learning (AutoML) that uses heterogeneous techniques. In contrast with existing AutoML approaches, our contribution can automatically build machine learning pipelines that combine techniques and algorithms from different frameworks, including shallow classifiers, natural language processing tools, and neural networks. We define the heterogeneous AutoML optimization problem as the search for the best sequence of algorithms that transforms specific input data into the desired output. 
This provides a novel theoretical and practical approach to AutoML. Our proposal is experimentally evaluated in diverse machine learning problems and compared with alternative approaches, showing that it is competitive with other AutoML alternatives in standard benchmarks. Furthermore, it can be applied to novel scenarios, such as several NLP tasks, where existing alternatives cannot be
directly deployed. The system is freely available and includes in-built compatibility with a large number of popular machine learning frameworks, which makes our approach useful for solving practical problems with relative ease and effort.

Downloads

Next from COLING 2020

Better Sign Language Translation with STMC-Transformer

Similar lecture

German’s Next Language Model

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from COLING 2020

Better Sign Language Translation with STMC-Transformer

Similar lecture

German’s Next Language Model

Downloads