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

TravelBERT: Pre-training Language Model Incorporating Domain-specific Heterogeneous Knowledge into A Unified Representation

Sep 05, 2021
Hongyin Zhu, Hao Peng, Zhiheng Lyu, Lei Hou, Juanzi Li, Jinghui Xiao

Existing technologies expand BERT from different perspectives, e.g. designing different pre-training tasks, different semantic granularities and different model architectures. Few models consider expanding BERT from different text formats. In this paper, we propose a heterogeneous knowledge language model (HKLM), a unified pre-trained language model (PLM) for all forms of text, including unstructured text, semi-structured text and well-structured text. To capture the corresponding relations among these multi-format knowledge, our approach uses masked language model objective to learn word knowledge, uses triple classification objective and title matching objective to learn entity knowledge and topic knowledge respectively. To obtain the aforementioned multi-format text, we construct a corpus in the tourism domain and conduct experiments on 5 tourism NLP datasets. The results show that our approach outperforms the pre-training of plain text using only 1/4 of the data. The code, datasets, corpus and knowledge graph will be released.

  Access Paper or Ask Questions

Simultaneous Multiple-Prompt Guided Generation Using Differentiable Optimal Transport

Apr 18, 2022
Yingtao Tian, Marco Cuturi, David Ha

Recent advances in deep learning, such as powerful generative models and joint text-image embeddings, have provided the computational creativity community with new tools, opening new perspectives for artistic pursuits. Text-to-image synthesis approaches that operate by generating images from text cues provide a case in point. These images are generated with a latent vector that is progressively refined to agree with text cues. To do so, patches are sampled within the generated image, and compared with the text prompts in the common text-image embedding space; The latent vector is then updated, using gradient descent, to reduce the mean (average) distance between these patches and text cues. While this approach provides artists with ample freedom to customize the overall appearance of images, through their choice in generative models, the reliance on a simple criterion (mean of distances) often causes mode collapse: The entire image is drawn to the average of all text cues, thereby losing their diversity. To address this issue, we propose using matching techniques found in the optimal transport (OT) literature, resulting in images that are able to reflect faithfully a wide diversity of prompts. We provide numerous illustrations showing that OT avoids some of the pitfalls arising from estimating vectors with mean distances, and demonstrate the capacity of our proposed method to perform better in experiments, qualitatively and quantitatively.

* Accepted at ICCC 2022 

  Access Paper or Ask Questions

The meta book and size-dependent properties of written language

Sep 24, 2009
Sebastian Bernhardsson, Luis Enrique Correa da Rocha, Petter Minnhagen

Evidence is given for a systematic text-length dependence of the power-law index gamma of a single book. The estimated gamma values are consistent with a monotonic decrease from 2 to 1 with increasing length of a text. A direct connection to an extended Heap's law is explored. The infinite book limit is, as a consequence, proposed to be given by gamma = 1 instead of the value gamma=2 expected if the Zipf's law was ubiquitously applicable. In addition we explore the idea that the systematic text-length dependence can be described by a meta book concept, which is an abstract representation reflecting the word-frequency structure of a text. According to this concept the word-frequency distribution of a text, with a certain length written by a single author, has the same characteristics as a text of the same length pulled out from an imaginary complete infinite corpus written by the same author.

* New J. Phys. 11 (2009) 123015 
* 7 pages, 6 figures, 1 table 

  Access Paper or Ask Questions

Efficient Model-Based Multi-Agent Mean-Field Reinforcement Learning

Jul 08, 2021
Barna Pasztor, Ilija Bogunovic, Andreas Krause

Learning in multi-agent systems is highly challenging due to the inherent complexity introduced by agents' interactions. We tackle systems with a huge population of interacting agents (e.g., swarms) via Mean-Field Control (MFC). MFC considers an asymptotically infinite population of identical agents that aim to collaboratively maximize the collective reward. Specifically, we consider the case of unknown system dynamics where the goal is to simultaneously optimize for the rewards and learn from experience. We propose an efficient model-based reinforcement learning algorithm $\text{M}^3\text{-UCRL}$ that runs in episodes and provably solves this problem. $\text{M}^3\text{-UCRL}$ uses upper-confidence bounds to balance exploration and exploitation during policy learning. Our main theoretical contributions are the first general regret bounds for model-based RL for MFC, obtained via a novel mean-field type analysis. $\text{M}^3\text{-UCRL}$ can be instantiated with different models such as neural networks or Gaussian Processes, and effectively combined with neural network policy learning. We empirically demonstrate the convergence of $\text{M}^3\text{-UCRL}$ on the swarm motion problem of controlling an infinite population of agents seeking to maximize location-dependent reward and avoid congested areas.

* 28 pages, 2 figures, Preprint, Submitted to NeurIPS 2021 

  Access Paper or Ask Questions

Joint Intermodal and Intramodal Label Transfers for Extremely Rare or Unseen Classes

Mar 22, 2017
Guo-Jun Qi, Wei Liu, Charu Aggarwal, Thomas Huang

In this paper, we present a label transfer model from texts to images for image classification tasks. The problem of image classification is often much more challenging than text classification. On one hand, labeled text data is more widely available than the labeled images for classification tasks. On the other hand, text data tends to have natural semantic interpretability, and they are often more directly related to class labels. On the contrary, the image features are not directly related to concepts inherent in class labels. One of our goals in this paper is to develop a model for revealing the functional relationships between text and image features as to directly transfer intermodal and intramodal labels to annotate the images. This is implemented by learning a transfer function as a bridge to propagate the labels between two multimodal spaces. However, the intermodal label transfers could be undermined by blindly transferring the labels of noisy texts to annotate images. To mitigate this problem, we present an intramodal label transfer process, which complements the intermodal label transfer by transferring the image labels instead when relevant text is absent from the source corpus. In addition, we generalize the inter-modal label transfer to zero-shot learning scenario where there are only text examples available to label unseen classes of images without any positive image examples. We evaluate our algorithm on an image classification task and show the effectiveness with respect to the other compared algorithms.

* The paper has been accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence. It will apear in a future issue 

  Access Paper or Ask Questions

Language Networks: a Practical Approach

Oct 13, 2020
Jorge A. V. Tohalino, Diego R. Amancio

This manuscript provides a short and practical introduction to the topic of language networks. This text aims at assisting researchers with no practical experience in text and/or network analysis. We provide a practical tutorial on how to model and characterize texts using network-based features. In this tutorial, we also include examples of pre-processing and network representations. A brief description of the main tasks allying network science and text analysis is also provided. A further development of this text shall include a practical description of network classification via machine learning methods.

  Access Paper or Ask Questions

What If We Only Use Real Datasets for Scene Text Recognition? Toward Scene Text Recognition With Fewer Labels

Mar 07, 2021
Jeonghun Baek, Yusuke Matsui, Kiyoharu Aizawa

Scene text recognition (STR) task has a common practice: All state-of-the-art STR models are trained on large synthetic data. In contrast to this practice, training STR models only on fewer real labels (STR with fewer labels) is important when we have to train STR models without synthetic data: for handwritten or artistic texts that are difficult to generate synthetically and for languages other than English for which we do not always have synthetic data. However, there has been implicit common knowledge that training STR models on real data is nearly impossible because real data is insufficient. We consider that this common knowledge has obstructed the study of STR with fewer labels. In this work, we would like to reactivate STR with fewer labels by disproving the common knowledge. We consolidate recently accumulated public real data and show that we can train STR models satisfactorily only with real labeled data. Subsequently, we find simple data augmentation to fully exploit real data. Furthermore, we improve the models by collecting unlabeled data and introducing semi- and self-supervised methods. As a result, we obtain a competitive model to state-of-the-art methods. To the best of our knowledge, this is the first study that 1) shows sufficient performance by only using real labels and 2) introduces semi- and self-supervised methods into STR with fewer labels. Our code and data are available:

* CVPR 2021 

  Access Paper or Ask Questions

Self-Supervised Learning from Web Data for Multimodal Retrieval

Jan 07, 2019
Raul Gomez, Lluis Gomez, Jaume Gibert, Dimosthenis Karatzas

Self-Supervised learning from multimodal image and text data allows deep neural networks to learn powerful features with no need of human annotated data. Web and Social Media platforms provide a virtually unlimited amount of this multimodal data. In this work we propose to exploit this free available data to learn a multimodal image and text embedding, aiming to leverage the semantic knowledge learnt in the text domain and transfer it to a visual model for semantic image retrieval. We demonstrate that the proposed pipeline can learn from images with associated textwithout supervision and analyze the semantic structure of the learnt joint image and text embedding space. We perform a thorough analysis and performance comparison of five different state of the art text embeddings in three different benchmarks. We show that the embeddings learnt with Web and Social Media data have competitive performances over supervised methods in the text based image retrieval task, and we clearly outperform state of the art in the MIRFlickr dataset when training in the target data. Further, we demonstrate how semantic multimodal image retrieval can be performed using the learnt embeddings, going beyond classical instance-level retrieval problems. Finally, we present a new dataset, InstaCities1M, composed by Instagram images and their associated texts that can be used for fair comparison of image-text embeddings.

* Submitted to Multi-Modal Scene Understanding. arXiv admin note: substantial text overlap with arXiv:1808.06368 

  Access Paper or Ask Questions

Approaching the linguistic complexity

Jan 21, 2009
Stanislaw Drozdz, Jaroslaw Kwapien, Adam Orczyk

We analyze the rank-frequency distributions of words in selected English and Polish texts. We compare scaling properties of these distributions in both languages. We also study a few small corpora of Polish literary texts and find that for a corpus consisting of texts written by different authors the basic scaling regime is broken more strongly than in the case of comparable corpus consisting of texts written by the same author. Similarly, for a corpus consisting of texts translated into Polish from other languages the scaling regime is broken more strongly than for a comparable corpus of native Polish texts. Moreover, based on the British National Corpus, we consider the rank-frequency distributions of the grammatically basic forms of words (lemmas) tagged with their proper part of speech. We find that these distributions do not scale if each part of speech is analyzed separately. The only part of speech that independently develops a trace of scaling is verbs.

* Complex Sciences, Lect. Notes ICST vol.4, 1044-1050 (Springer, 2009) 
* to be published in conference proceedings 

  Access Paper or Ask Questions

Predicting Actions to Help Predict Translations

Aug 18, 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