Spain

Transformer becomes the state-of-the-art translation model, while it is not well studied how each intermediate component contributes to the model performance, which poses significant challenges for designing optimal architectures. In this work, we bridge this gap by evaluating the impact of individual component (sub-layer) in trained Transformer models from different perspectives. Experimental results across language pairs, training strategies, and model capacities show that certain components are consistently more important than the others. We also report a number of interesting findings that might help humans better analyze, understand and improve Transformer models. Based on these observations, we further propose a new training strategy that can improves translation performance by distinguishing the unimportant components in training.

COLING 2020

Rethinking the Value of Transformer Components

interpretability

machine translation

transformer

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

Visualization and topic modeling are widely used approaches for text analysis. Traditional visualization methods find low-dimensional representations of documents in the visualization space (typically 2D or 3D) that can be displayed using a scatterplot. In contrast, topic modeling aims to discover topics from text, but for visualization, one needs to perform a post-hoc embedding using dimensionality reduction methods. Recent approaches propose using a generative model to jointly find topics and visualization, allowing the semantics to be infused in the visualization
space for a meaningful interpretation. A major challenge that prevents these methods from being used practically is the scalability of their inference algorithms. We present, to the best of our knowledge, the first fast Auto-Encoding Variational Bayes based inference method for jointly inferring topics and visualization. Since our method is black box, it can handle model changes efficiently with little mathematical rederivation effort. We demonstrate the efficiency and effectiveness of our method on real-world large datasets and compare it with existing baselines.

Auto-Encoding Variational Bayes for Inferring Topics and Visualization

Text-to-SQL systems offers natural language interfaces to databases, which can automatically generates SQL queries given natural language questions. On the WikiSQL benchmark, state-ofthe-art text-to-SQL systems typically take a slot-filling approach by building several specialized models for each type of slot. Despite being effective, such modularized systems are complex and also fall short in jointly learning for different slots. To solve these problems, this paper proposes a novel approach that formulates the task as a question answering problem, where different slots are predicted by a unified machine reading comprehension (MRC) model. For this purpose, we use a BERT-based MRC model, which can also benefit from intermediate training on other MRC datasets. The proposed method can achieve competitive results on WikiSQL, suggesting it being a promising direction for text-to-SQL.

SQL Generation via Machine Reading Comprehension

This paper proposed a supervised visual attention mechanism for multimodal neural machine translation (MNMT), trained with constraints based on manual alignments between words in a sentence and their corresponding regions of an image. 
The proposed visual attention mechanism captures the relationship between a word and an image region more precisely than a conventional visual attention mechanism trained through MNMT in an unsupervised manner. 
Our experiments on English-German and German-English translation tasks using the Multi30k dataset and on English-Japanese and Japanese-English translation tasks using the Flickr30k Entities JP dataset show that a Transformer-based MNMT model can be improved by incorporating our proposed supervised visual attention mechanism and that further improvements can be achieved by combining it with a supervised cross-lingual attention mechanism (up to +1.61 BLEU, +1.7 METEOR).

Supervised Visual Attention for Multimodal Neural Machine Translation

We present a large-scale corpus of e-mail conversations with domain-agnostic and two-level dialogue act (DA) annotations towards the goal of a better understanding of asynchronous conversations. To assess the difficulty of DA recognition on our corpus, we evaluate several models, including a pre-trained contextual representation model, as our baselines. The experimental results show that BERT outperforms other neural network models, including previous state-of-the-art models, but falls short of a human performance. An evaluation of a model trained on our corpus against other domains of asynchronous conversation reveals the domain independence of our DA annotations.

A Large-Scale Corpus of E-mail Conversations with Standard and Two-Level Dialogue Act Annotations

Multi-turn response selection has been extensively studied and applied to many real-world applications in recent years. However, current methods typically model the interactions between multi-turn utterances and candidate responses with iterative approaches, which is not practical as the turns of conversations vary. Besides, some latent features, such as user intent and conversation topic, are under-discovered in existing works. In this work, we propose Intra-/Inter-Interaction Network (I3) with latent interaction modeling to comprehensively model multi-level interactions between the utterance context and the response. In specific, we first encode the intra- and inter-utterance interaction with the given response from both individual utterance and the overall utterance context. Then we develop a latent multi-view subspace clustering module to model the latent interaction between the utterance and response. Experimental results show that the proposed method substantially and consistently outperforms existing state-of-the-art methods on three multi-turn response selection benchmark datasets.

Intra-/Inter-Interaction Network with Latent Interaction Modeling for Multi-turn Response Selection

We wish to explore the problem of discovering conversational intent from past human-human conversation dataset. We assume task-oriented conversational intent where one user is trying to accomplish and the agent is helping to solve that problem. We extract short text representation of a conversation using a novel feature extraction pipeline. By contrast, we formulate the problem of conversational intent mining as a short text clustering. We found that the conversational intents are unbalanced in nature. Since density based clustering can find clusters of arbitrary shapes, we investigate popular algorithms such as DBSCAN, HDBSCAN for unbalanced data clustering problem. We also propose a simple formulation of the density based clustering algorithm (ITER-DBSCAN) and propose a parallel version for large dataset. Our algorithm achieves state-of-the art results on six intent mining, one conversational dataset and one short text clustering dataset. Codes and dataset are published on GitHub for future research.

Intent Mining from past conversations for Conversational Agent

This paper introduces a method that efficiently reduces the computational cost and parameter size of Transformer. The proposed model, refer to as \textit{Group-Transformer}, splits feature space into multiple groups, factorizes the calculation paths, and reduces computations for the group interaction.
Extensive experiments on two benchmark tasks, enwik8 and text8, prove our model's effectiveness and efficiency in small-scale Transformers.
To the best of our knowledge, Group-Transformer is the first attempt to design Transformer with the group strategy, widely used for efficient CNN architectures.

Scale down Transformer by Grouping Features for a Lightweight Character-level Language Model

In this paper, we separate the model into a sentence reconstruction module and a style module. We use the Transformer-based autoencoder model for sentence reconstruction and the adaptive style embedding is learned directly in the style module. Because of this separation, each module can better focus on its own task. Moreover, we can vary the style strength of the generated sentence by changing the style of the embedding expression.

How Positive Are You:Text Style Transfer using Adaptive Style Embedding

We propose a distance-based model aiming at temporal knowledge graph embedding (TKGE). The temporal evolution of an entity embedding is defined as a rotation from the initial time to the current time in the complex vector space. Each relation is represented as a pair of dual complex embeddings to handle the beginning and the end of the relation, respectively. We use different time division methods and time granularities for different datasets. Experimental Results show that our model achieves the state-of-the-art performances on four TKG datasets.


TeRo: A Time-aware Knowledge Graph Embedding Model via Temporal Rotation

Lexical substitution, i.e. generation of plausible words that can replace a particular target word in a given context, is an extremely powerful technology that can be used as a backbone of various NLP applications, including word sense induction and disambiguation, lexical relation extraction, data augmentation, etc. In this paper, we present a large-scale comparative study of lexical substitution methods employing both rather old and most recent language and masked language models (LMs and MLMs), such as context2vec, ELMo, BERT, RoBERTa, XLNet. We show that already competitive results achieved by SOTA LMs/MLMs can be further substantially improved if information about the target word is injected properly. Several existing and new target word injection methods are compared for each LM/MLM using both intrinsic evaluation on lexical substitution datasets and extrinsic evaluation on word sense induction (WSI) datasets. On two WSI datasets we obtain new SOTA results. Besides, we analyze the types of semantic relations between target words and their substitutes generated by different models or given by annotators.

Rethinking the Value of Transformer Components

Downloads

Next from COLING 2020

Auto-Encoding Variational Bayes for Inferring Topics and Visualization

Similar lecture

How Far Does BERT Look At: Distance-based Clustering and Analysis of BERT's Attention

Rethinking the Value of Transformer Components

.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

Auto-Encoding Variational Bayes for Inferring Topics and Visualization

Similar lecture

How Far Does BERT Look At: Distance-based Clustering and Analysis of BERT's Attention

Downloads