Spain

How do people understand concepts such as olive oil, baby oil, or lamp oil? Embodied approaches to knowledge representation suggest that words are represented as a set of features that are the basic components of meaning. In particular, Binder et al. (2009) grounded this idea by mapping semantic features (attributes) to different brain systems in their Concept Attribute Representations (CAR) theory. fMRI experiments demonstrate that when humans listen or read sentences they use different brain systems to simulate seeing the scenes and performing the actions that are described. An intriguing challenge to this theory is that concepts are dynamic, i.e. word meaning depends on context. This work addresses such challenge through the Context-dEpendent meaning REpresentations in the BRAin (CEREBRA) neural network model. Based on changes in the fMRI patterns, CEREBRA quantified how word meanings change in the context of a sentence. CEREBRA is evaluated through different computational experiments and a behavioral analysis. The experiments demonstrated that words in different contexts have different representations, the changes observed in the concept attributes revealed unique conceptual combinations, and that the new representations were more similar to the other words in the sentence than to the original representations. The behavioral analysis confirmed that the changes produced by CEREBRA are actionable knowledge that can be used to predict human responses. Together, these experiments constitute a comprehensive evaluation of CEREBRA’s context-based representations as well as the soundness of CAR theory. Thus, CEREBRA is a useful tool for understanding how semantic knowledge is represented in the brain, and for providing a human-like context-based representations for NLP applications.

COLING 2020

Characterizing Dynamic Word Meaning Representations in the Brain

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.

technical paper

Understanding context-dependent variation in word meanings is a key aspect of human language comprehension supported by the lexicon. Lexicographic resources (e.g., WordNet) capture only some of this context-dependent variation; for example, they often do not encode how closely senses, or discretized word meanings, are related to one another. Our work investigates whether recent advances in NLP, specifically contextualized word embeddings, capture human-like distinctions between English word senses, such as polysemy and homonymy. We collect data from a behavioral, web-based experiment, in which participants provide judgments of the relatedness of multiple WordNet senses of a word in a two-dimensional spatial arrangement task. We find that participants' judgments of the relatedness between senses are correlated with distances between senses in the BERT embedding space. Homonymous senses (e.g., bat as mammal vs. bat as sports equipment) are reliably more distant from one another in the embedding space than polysemous ones (e.g., chicken as animal vs. chicken as meat). Our findings point towards the potential utility of continuous-space representations of sense meanings.

Contextualized Word Embeddings Encode Aspects of Human-Like Word Sense Knowledge

Word Association Norms (WAN) are collections that present stimuli words and the set of their associated responses. The corpus is widely used in diverse areas of expertise. In order to reduce the effort to have a good quality resource that can be reproduced in many languages with minimum sources, a methodology to build AutomaticWord Association Norms is proposed (AWAN).
The methodology has an input of two simple elements: a) dictionary, and b) pre-processed Word Embeddings. This new kind of WAN is evaluated in two ways: i) learning word embeddings based on the node2vec algorithm and comparing them with human annotated benchmarks, and ii) performing a lexical search for a reverse dictionary. Both evaluations are done in a weighted graph with the AWAN lexical elements. The results showed that the methodology produces good quality AWANs.

AutomaticWord Association Norms (AWAN)

Knowledge graphs (KGs) of real-world facts about entities and their relationships are useful resources for a variety of natural language processing tasks. However, because knowledge graphs are typically incomplete, it is useful to perform knowledge graph completion or link prediction, i.e. predict whether a relationship not in the knowledge graph is likely to be true. This paper serves as a comprehensive survey of embedding models of entities and relationships for knowledge graph completion, summarizing up-to-date experimental results on standard benchmark datasets and pointing out potential future research directions.

A survey of embedding models of entities and relationships for knowledge graph completion

We propose the Graph2Graph Transformer architecture for conditioning on and predicting arbitrary graphs, and apply it to the challenging task of transition-based dependency parsing. After proposing two novel Transformer models of transition-based dependency parsing as strong baselines, we show that adding the proposed mechanisms for conditioning on and predicting graphs of Graph2Graph Transformer results in significant improvements, both with and without BERT pre-training. The novel baselines and their integration with Graph2Graph Transformer significantly outperform the state-of-the-art in traditional transition-based dependency parsing on both English Penn Treebank, and 13 languages of Universal Dependencies Treebanks. Graph2Graph Transformer can be integrated with many previous structured prediction methods, making it easy to apply to a wide range of NLP tasks.

Graph-to-Graph Transformer for Transition-based Dependency Parsing

Entity Resolution (ER) identifies records that refer to the same real-world entity. Deep learning approaches improved the generalization ability of entity matching models, but hardly overcame the impact of noisy or incomplete data sources. In real scenes, an entity usually consists of multiple semantic facets, called aspects. In this paper, we focus on entity augmentation, namely retrieving the values of missing aspects. The relationship between aspects is naturally suitable to be represented by a knowledge graph, where entity augmentation can be modeled as a link prediction problem. Our paper proposes a novel graph-based approach to solve entity augmentation. Specifically, we apply a dedicated random walk algorithm, which uses node types to limit the traversal length, and encodes graph structure into low-dimensional embeddings. Thus, the missing aspects could be retrieved by a link prediction model. Furthermore, the augmented aspects with fixed orders are served as the input of a deep Siamese BiLSTM network for entity matching. We compared our method with state-of-the-art methods through extensive experiments on downstream ER tasks. According to the experiment results, our model outperforms other methods on evaluation metrics (accuracy, precision, recall, and f1-score) to a large extent, which demonstrates the effectiveness of our method.

Graph-based Aspect Representation Learning for Entity Resolution

We propose a simple and efficient framework to learn syntactic embeddings based on information derived from constituency parse trees. Using biased random walk methods, our embeddings not only encode syntactic information about words, but they also capture contextual information. We also propose a method to train the embeddings on multiple constituency parse trees to ensure the encoding of global syntactic representation. Quantitative evaluation of the embeddings shows competitive performance on POS tagging task when compared to other types of embeddings, and qualitative evaluation reveals interesting facts about the syntactic typology learned by these embeddings.

Graph-based Syntactic Word Embeddings

Named entity recognition (NER) from visual documents, such as invoices, receipts or business cards, is a critical task for visual document understanding. Most classical approaches use a sequence-based model (typically BiLSTM-CRF framework) without considering document structure. Recent work on graph-based model using graph convolutional networks to encode visual and textual features have achieved promising performance on the task. However, few attempts take geometry information of text segments (text in bounding box) in visual documents into account. Meanwhile, existing methods do not consider that related text segments which need to be merged to form a complete entity in many real-world situations. In this paper, we present GraphNEMR, a graph-based model that uses graph convolutional networks to jointly merge text segments and recognize named entities. By incorporating geometry information from visual documents into our model, richer 2D context information is generated to improve document representations. To merge text segments, we introduce a novel mechanism that captures both geometry information as well as semantic information based on pre-trained language model. Experimental results show that the proposed GraphNEMR model outperforms both sequence-based and graph-based SOTA methods significantly.

Merge and Recognize: A Geometry and 2D Context Aware Graph Model for Named Entity Recognition from Visual Documents

Graph-based semi-supervised learning is appealing when labels are scarce but large amounts of
unlabeled data are available. These methods typically use a heuristic strategy to construct the
graph based on some fixed data representation, independently of the available labels. In this pa-
per, we propose to jointly learn a data representation and a graph from both labeled and unlabeled
data such that (i) the learned representation indirectly encodes the label information injected into
the graph, and (ii) the graph provides a smooth topology with respect to the transformed data.
Plugging the resulting graph and representation into existing graph-based semi-supervised learn-
ing algorithms like label spreading and graph convolutional networks, we show that our approach
outperforms standard graph construction methods on both synthetic data and real datasets.

Joint Learning of the Graph and the Data Representation for Graph-Based Semi-Supervised Learning

BERT is a popular language model whose main pre-training task is to fill in the blank, i.e., predicting a word that was masked out of a sentence, based on the remaining words. In some applications, however, having an additional context can help the model make the right prediction, e.g., by taking the domain or the time of writing into account. This motivates us to advance the BERT architecture by adding a global state for conditioning on a fixed-sized context. We present our two novel approaches and apply them to an industry use-case, where we complete fashion outfits with missing articles, conditioned on a specific customer. An experimental comparison to other methods from the literature shows that our methods improve personalization significantly.

Contextual BERT: Conditioning the Language Model Using a Global State

We propose an open-world knowledge graph completion model that can be combined with common closed-world approaches (such as ComplEx) and enhance them to exploit text-based representations for entities unseen in training. Our model learns relation-specific transformation functions from text-based to graph-based embedding space, where the closed-world link prediction model can be applied. We demonstrate state-of-the-art results on common open-world benchmarks and show that our approach benefits from relation-specific transformation functions (RST), giving substantial improvements over a relation-agnostic approach.

Characterizing Dynamic Word Meaning Representations in the Brain

Downloads

Next from COLING 2020

Contextualized Word Embeddings Encode Aspects of Human-Like Word Sense Knowledge

Similar lecture

Individual corpora predict fast memory retrieval during reading

Characterizing Dynamic Word Meaning Representations in the Brain

.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

Contextualized Word Embeddings Encode Aspects of Human-Like Word Sense Knowledge

Similar lecture

Individual corpora predict fast memory retrieval during reading

Downloads