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

RareAct: A video dataset of unusual interactions

Aug 03, 2020
Antoine Miech, Jean-Baptiste Alayrac, Ivan Laptev, Josef Sivic, Andrew Zisserman

This paper introduces a manually annotated video dataset of unusual actions, namely RareAct, including actions such as "blend phone", "cut keyboard" and "microwave shoes". RareAct aims at evaluating the zero-shot and few-shot compositionality of action recognition models for unlikely compositions of common action verbs and object nouns. It contains 122 different actions which were obtained by combining verbs and nouns rarely co-occurring together in the large-scale textual corpus from HowTo100M, but that frequently appear separately. We provide benchmarks using a state-of-the-art HowTo100M pretrained video and text model and show that zero-shot and few-shot compositionality of actions remains a challenging and unsolved task.

  Access Paper or Ask Questions

Clue: Cross-modal Coherence Modeling for Caption Generation

May 02, 2020
Malihe Alikhani, Piyush Sharma, Shengjie Li, Radu Soricut, Matthew Stone

We use coherence relations inspired by computational models of discourse to study the information needs and goals of image captioning. Using an annotation protocol specifically devised for capturing image--caption coherence relations, we annotate 10,000 instances from publicly-available image--caption pairs. We introduce a new task for learning inferences in imagery and text, coherence relation prediction, and show that these coherence annotations can be exploited to learn relation classifiers as an intermediary step, and also train coherence-aware, controllable image captioning models. The results show a dramatic improvement in the consistency and quality of the generated captions with respect to information needs specified via coherence relations.

* Accepted as a long paper to ACL 2020 

  Access Paper or Ask Questions

Automated Content Grading Using Machine Learning

Apr 08, 2020
Rahul Kr Chauhan, Ravinder Saharan, Siddhartha Singh, Priti Sharma

Grading of examination papers is a hectic, time-labor intensive task and is often subjected to inefficiency and bias in checking. This research project is a primitive experiment in the automation of grading of theoretical answers written in exams by students in technical courses which yet had continued to be human graded. In this paper, we show how the algorithmic approach in machine learning can be used to automatically examine and grade theoretical content in exam answer papers. Bag of words, their vectors & centroids, and a few semantic and lexical text features have been used overall. Machine learning models have been implemented on datasets manually built from exams given by graduating students enrolled in technical courses. These models have been compared to show the effectiveness of each model.

* 7 pages, 6 figures 

  Access Paper or Ask Questions

How Furiously Can Colourless Green Ideas Sleep? Sentence Acceptability in Context

Apr 02, 2020
Jey Han Lau, Carlos S. Armendariz, Shalom Lappin, Matthew Purver, Chang Shu

We study the influence of context on sentence acceptability. First we compare the acceptability ratings of sentences judged in isolation, with a relevant context, and with an irrelevant context. Our results show that context induces a cognitive load for humans, which compresses the distribution of ratings. Moreover, in relevant contexts we observe a discourse coherence effect which uniformly raises acceptability. Next, we test unidirectional and bidirectional language models in their ability to predict acceptability ratings. The bidirectional models show very promising results, with the best model achieving a new state-of-the-art for unsupervised acceptability prediction. The two sets of experiments provide insights into the cognitive aspects of sentence processing and central issues in the computational modelling of text and discourse.

* 14 pages. Author's final version, accepted for publication in Transactions of the Association for Computational Linguistics 

  Access Paper or Ask Questions

A Multi-Source Entity-Level Sentiment Corpus for the Financial Domain: The FinLin Corpus

Mar 09, 2020
Tobias Daudert

We introduce FinLin, a novel corpus containing investor reports, company reports, news articles, and microblogs from StockTwits, targeting multiple entities stemming from the automobile industry and covering a 3-month period. FinLin was annotated with a sentiment score and a relevance score in the range [-1.0, 1.0] and [0.0, 1.0], respectively. The annotations also include the text spans selected for the sentiment, thus, providing additional insight into the annotators' reasoning. Overall, FinLin aims to complement the current knowledge by providing a novel and publicly available financial sentiment corpus and to foster research on the topic of financial sentiment analysis and potential applications in behavioural science.

  Access Paper or Ask Questions

Dynamic Prosody Generation for Speech Synthesis using Linguistics-Driven Acoustic Embedding Selection

Dec 02, 2019
Shubhi Tyagi, Marco Nicolis, Jonas Rohnke, Thomas Drugman, Jaime Lorenzo-Trueba

Recent advances in Text-to-Speech (TTS) have improved quality and naturalness to near-human capabilities when considering isolated sentences. But something which is still lacking in order to achieve human-like communication is the dynamic variations and adaptability of human speech. This work attempts to solve the problem of achieving a more dynamic and natural intonation in TTS systems, particularly for stylistic speech such as the newscaster speaking style. We propose a novel embedding selection approach which exploits linguistic information, leveraging the speech variability present in the training dataset. We analyze the contribution of both semantic and syntactic features. Our results show that the approach improves the prosody and naturalness for complex utterances as well as in Long Form Reading (LFR).

* Submitted for ICASSP 2020 

  Access Paper or Ask Questions

High Quality ELMo Embeddings for Seven Less-Resourced Languages

Nov 22, 2019
Matej Ulčar, Marko Robnik-Šikonja

Recent results show that deep neural networks using contextual embeddings significantly outperform non-contextual embeddings on a majority of text classification task. We offer precomputed embeddings from popular contextual ELMo model for seven languages: Croatian, Estonian, Finnish, Latvian, Lithuanian, Slovenian, and Swedish. We demonstrate that the quality of embeddings strongly depends on the size of training set and show that existing publicly available ELMo embeddings for listed languages shall be improved. We train new ELMo embeddings on much larger training sets and show their advantage over baseline non-contextual FastText embeddings. In evaluation, we use two benchmarks, the analogy task and the NER task.

* 5 pages, 1 figure 

  Access Paper or Ask Questions

Meta Answering for Machine Reading

Nov 11, 2019
Benjamin Borschinger, Jordan Boyd-Graber, Christian Buck, Jannis Bulian, Massimiliano Ciaramita, Michelle Chen Huebscher, Wojciech Gajewski, Yannic Kilcher, Rodrigo Nogueira, Lierni Sestorain Saralegu

We investigate a framework for machine reading, inspired by real world information-seeking problems, where a meta question answering system interacts with a black box environment. The environment encapsulates a competitive machine reader based on BERT, providing candidate answers to questions, and possibly some context. To validate the realism of our formulation, we ask humans to play the role of a meta-answerer. With just a small snippet of text around an answer, humans can outperform the machine reader, improving recall. Similarly, a simple machine meta-answerer outperforms the environment, improving both precision and recall on the Natural Questions dataset. The system relies on joint training of answer scoring and the selection of conditioning information.

  Access Paper or Ask Questions

Relative contributions of Shakespeare and Fletcher in Henry VIII: An Analysis Based on Most Frequent Words and Most Frequent Rhythmic Patterns

Oct 30, 2019
Petr Plecháč

The versified play Henry VIII is nowadays widely recognized to be a collaborative work not written solely by William Shakespeare. We employ combined analysis of vocabulary and versification together with machine learning techniques to determine which authors also took part in the writing of the play and what were their relative contributions. Unlike most previous studies, we go beyond the attribution of particular scenes and use the rolling attribution approach to determine the probabilities of authorship of pieces of texts, without respecting the scene boundaries. Our results highly support the canonical division of the play between William Shakespeare and John Fletcher proposed by James Spedding, but also bring new evidence supporting the modifications proposed later by Thomas Merriam.

  Access Paper or Ask Questions

MinWikiSplit: A Sentence Splitting Corpus with Minimal Propositions

Sep 26, 2019
Christina Niklaus, Andre Freitas, Siegfried Handschuh

We compiled a new sentence splitting corpus that is composed of 203K pairs of aligned complex source and simplified target sentences. Contrary to previously proposed text simplification corpora, which contain only a small number of split examples, we present a dataset where each input sentence is broken down into a set of minimal propositions, i.e. a sequence of sound, self-contained utterances with each of them presenting a minimal semantic unit that cannot be further decomposed into meaningful propositions. This corpus is useful for developing sentence splitting approaches that learn how to transform sentences with a complex linguistic structure into a fine-grained representation of short sentences that present a simple and more regular structure which is easier to process for downstream applications and thus facilitates and improves their performance.

  Access Paper or Ask Questions