Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

"Text": models, code, and papers

Predicting Actions to Help Predict Translations

Aug 05, 2019
Zixiu Wu, Julia Ive, Josiah Wang, Pranava Madhyastha, Lucia Specia

We address the task of text translation on the How2 dataset using a state of the art transformer-based multimodal approach. The question we ask ourselves is whether visual features can support the translation process, in particular, given that this is a dataset extracted from videos, we focus on the translation of actions, which we believe are poorly captured in current static image-text datasets currently used for multimodal translation. For that purpose, we extract different types of action features from the videos and carefully investigate how helpful this visual information is by testing whether it can increase translation quality when used in conjunction with (i) the original text and (ii) the original text where action-related words (or all verbs) are masked out. The latter is a simulation that helps us assess the utility of the image in cases where the text does not provide enough context about the action, or in the presence of noise in the input text.

* Accepted to workshop "The How2 Challenge: New Tasks for Vision & Language" of International Conference on Machine Learning 2019 

  Access Paper or Ask Questions

Tractability from overparametrization: The example of the negative perceptron

Oct 28, 2021
Andrea Montanari, Yiqiao Zhong, Kangjie Zhou

In the negative perceptron problem we are given $n$ data points $({\boldsymbol x}_i,y_i)$, where ${\boldsymbol x}_i$ is a $d$-dimensional vector and $y_i\in\{+1,-1\}$ is a binary label. The data are not linearly separable and hence we content ourselves to find a linear classifier with the largest possible \emph{negative} margin. In other words, we want to find a unit norm vector ${\boldsymbol \theta}$ that maximizes $\min_{i\le n}y_i\langle {\boldsymbol \theta},{\boldsymbol x}_i\rangle$. This is a non-convex optimization problem (it is equivalent to finding a maximum norm vector in a polytope), and we study its typical properties under two random models for the data. We consider the proportional asymptotics in which $n,d\to \infty$ with $n/d\to\delta$, and prove upper and lower bounds on the maximum margin $\kappa_{\text{s}}(\delta)$ or -- equivalently -- on its inverse function $\delta_{\text{s}}(\kappa)$. In other words, $\delta_{\text{s}}(\kappa)$ is the overparametrization threshold: for $n/d\le \delta_{\text{s}}(\kappa)-\varepsilon$ a classifier achieving vanishing training error exists with high probability, while for $n/d\ge \delta_{\text{s}}(\kappa)+\varepsilon$ it does not. Our bounds on $\delta_{\text{s}}(\kappa)$ match to the leading order as $\kappa\to -\infty$. We then analyze a linear programming algorithm to find a solution, and characterize the corresponding threshold $\delta_{\text{lin}}(\kappa)$. We observe a gap between the interpolation threshold $\delta_{\text{s}}(\kappa)$ and the linear programming threshold $\delta_{\text{lin}}(\kappa)$, raising the question of the behavior of other algorithms.

* 88 pages; 7 pdf figures 

  Access Paper or Ask Questions

Direct speech-to-speech translation with discrete units

Jul 12, 2021
Ann Lee, Peng-Jen Chen, Changhan Wang, Jiatao Gu, Xutai Ma, Adam Polyak, Yossi Adi, Qing He, Yun Tang, Juan Pino, Wei-Ning Hsu

We present a direct speech-to-speech translation (S2ST) model that translates speech from one language to speech in another language without relying on intermediate text generation. Previous work addresses the problem by training an attention-based sequence-to-sequence model that maps source speech spectrograms into target spectrograms. To tackle the challenge of modeling continuous spectrogram features of the target speech, we propose to predict the self-supervised discrete representations learned from an unlabeled speech corpus instead. When target text transcripts are available, we design a multitask learning framework with joint speech and text training that enables the model to generate dual mode output (speech and text) simultaneously in the same inference pass. Experiments on the Fisher Spanish-English dataset show that predicting discrete units and joint speech and text training improve model performance by 11 BLEU compared with a baseline that predicts spectrograms and bridges 83% of the performance gap towards a cascaded system. When trained without any text transcripts, our model achieves similar performance as a baseline that predicts spectrograms and is trained with text data.

  Access Paper or Ask Questions

Visual FUDGE: Form Understanding via Dynamic Graph Editing

May 17, 2021
Brian Davis, Bryan Morse, Brian Price, Chris Tensmeyer, Curtis Wiginton

We address the problem of form understanding: finding text entities and the relationships/links between them in form images. The proposed FUDGE model formulates this problem on a graph of text elements (the vertices) and uses a Graph Convolutional Network to predict changes to the graph. The initial vertices are detected text lines and do not necessarily correspond to the final text entities, which can span multiple lines. Also, initial edges contain many false-positive relationships. FUDGE edits the graph structure by combining text segments (graph vertices) and pruning edges in an iterative fashion to obtain the final text entities and relationships. While recent work in this area has focused on leveraging large-scale pre-trained Language Models (LM), FUDGE achieves the same level of entity linking performance on the FUNSD dataset by learning only visual features from the (small) provided training set. FUDGE can be applied on forms where text recognition is difficult (e.g. degraded or historical forms) and on forms in resource-poor languages where pre-training such LMs is challenging. FUDGE is state-of-the-art on the historical NAF dataset.

* Accepted at ICDAR 2021, 16 pages 

  Access Paper or Ask Questions

Dire n'est pas concevoir

Feb 10, 2010
Christophe Roche

The conceptual modelling built from text is rarely an ontology. As a matter of fact, such a conceptualization is corpus-dependent and does not offer the main properties we expect from ontology. Furthermore, ontology extracted from text in general does not match ontology defined by expert using a formal language. It is not surprising since ontology is an extra-linguistic conceptualization whereas knowledge extracted from text is the concern of textual linguistics. Incompleteness of text and using rhetorical figures, like ellipsis, modify the perception of the conceptualization we may have. Ontological knowledge, which is necessary for text understanding, is not in general embedded into documents.

* Ing\'enierie des Connaissances, Grenoble : France (2007) 
* 12 pages 

  Access Paper or Ask Questions

Diffusion Maps for Textual Network Embedding

May 24, 2018
Xinyuan Zhang, Yitong Li, Dinghan Shen, Lawrence Carin

Textual network embedding leverages rich text information associated with the network to learn low-dimensional vectorial representations of vertices. Rather than using typical natural language processing (NLP) approaches, recent research exploits the relationship of texts on the same edge to graphically embed text. However, these models neglect to measure the complete level of connectivity between any two texts in the graph. We present diffusion maps for textual network embedding (DMTE), integrating global structural information of the graph to capture the semantic relatedness between texts, with a diffusion-convolution operation applied on the text inputs. In addition, a new objective function is designed to efficiently preserve the high-order proximity using the graph diffusion. Experimental results show that the proposed approach outperforms state-of-the-art methods on the vertex-classification and link-prediction tasks.

  Access Paper or Ask Questions

Adversarial Robustness of Neural-Statistical Features in Detection of Generative Transformers

Mar 02, 2022
Evan Crothers, Nathalie Japkowicz, Herna Viktor, Paula Branco

The detection of computer-generated text is an area of rapidly increasing significance as nascent generative models allow for efficient creation of compelling human-like text, which may be abused for the purposes of spam, disinformation, phishing, or online influence campaigns. Past work has studied detection of current state-of-the-art models, but despite a developing threat landscape, there has been minimal analysis of the robustness of detection methods to adversarial attacks. To this end, we evaluate neural and non-neural approaches on their ability to detect computer-generated text, their robustness against text adversarial attacks, and the impact that successful adversarial attacks have on human judgement of text quality. We find that while statistical features underperform neural features, statistical features provide additional adversarial robustness that can be leveraged in ensemble detection models. In the process, we find that previously effective complex phrasal features for detection of computer-generated text hold little predictive power against contemporary generative models, and identify promising statistical features to use instead. Finally, we pioneer the usage of $\Delta$MAUVE as a proxy measure for human judgement of adversarial text quality.

  Access Paper or Ask Questions

Multiresolution Graph Attention Networks for Relevance Matching

Feb 27, 2019
Ting Zhang, Bang Liu, Di Niu, Kunfeng Lai, Yu Xu

A large number of deep learning models have been proposed for the text matching problem, which is at the core of various typical natural language processing (NLP) tasks. However, existing deep models are mainly designed for the semantic matching between a pair of short texts, such as paraphrase identification and question answering, and do not perform well on the task of relevance matching between short-long text pairs. This is partially due to the fact that the essential characteristics of short-long text matching have not been well considered in these deep models. More specifically, these methods fail to handle extreme length discrepancy between text pieces and neither can they fully characterize the underlying structural information in long text documents. In this paper, we are especially interested in relevance matching between a piece of short text and a long document, which is critical to problems like query-document matching in information retrieval and web searching. To extract the structural information of documents, an undirected graph is constructed, with each vertex representing a keyword and the weight of an edge indicating the degree of interaction between keywords. Based on the keyword graph, we further propose a Multiresolution Graph Attention Network to learn multi-layered representations of vertices through a Graph Convolutional Network (GCN), and then match the short text snippet with the graphical representation of the document with the attention mechanisms applied over each layer of the GCN. Experimental results on two datasets demonstrate that our graph approach outperforms other state-of-the-art deep matching models.

* Accepted by CIKM 2018 

  Access Paper or Ask Questions

The Harrington Yowlumne Narrative Corpus

Feb 01, 2021
Nathan M. White, Timothy Henry-Rodriguez

Minority languages continue to lack adequate resources for their development, especially in the technological domain. Likewise, the J.P. Harrington Papers collection at the Smithsonian Institution are difficult to access in practical terms for community members and researchers due to its handwritten and disorganized format. Our current work seeks to make a portion of this publicly-available yet problematic material practically accessible for natural language processing use. Here, we present the Harrington Yowlumne Narrative Corpus, a corpus of 20 narrative texts that derive from the Tejone\~no Yowlumne community of the Tinliw rancheria in Kern County, California between 1910 and 1925. We digitally transcribe the texts and provide gold-standard aligned lexeme-based normalized text with these texts. Altogether, the text contains 67,835 transcribed characters aligned with 10,721 gold standard text-normalized words.

  Access Paper or Ask Questions

Lexical Features Are More Vulnerable, Syntactic Features Have More Predictive Power

Sep 30, 2019
Jekaterina Novikova, Aparna Balagopalan, Ksenia Shkaruta, Frank Rudzicz

Understanding the vulnerability of linguistic features extracted from noisy text is important for both developing better health text classification models and for interpreting vulnerabilities of natural language models. In this paper, we investigate how generic language characteristics, such as syntax or the lexicon, are impacted by artificial text alterations. The vulnerability of features is analysed from two perspectives: (1) the level of feature value change, and (2) the level of change of feature predictive power as a result of text modifications. We show that lexical features are more sensitive to text modifications than syntactic ones. However, we also demonstrate that these smaller changes of syntactic features have a stronger influence on classification performance downstream, compared to the impact of changes to lexical features. Results are validated across three datasets representing different text-classification tasks, with different levels of lexical and syntactic complexity of both conversational and written language.

* EMNLP Workshop on Noisy User-generated Text (W-NUT 2019) 

  Access Paper or Ask Questions