We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers. Unlike recent language representation models, BERT is designed to pre-train deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers. As a result, the pre-trained BERT model can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks, such as question answering and language inference, without substantial task-specific architecture modifications. 
BERT is conceptually simple and empirically powerful. It obtains new state-of-the-art results on eleven natural language processing tasks, including pushing the GLUE score to 80.5% (7.7% point absolute improvement), MultiNLI accuracy to 86.7% (4.6% absolute improvement), SQuAD v1.1 question answering Test F1 to 93.2 (1.5 point absolute improvement) and SQuAD v2.0 Test F1 to 83.1 (5.1 point absolute improvement).

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers. Unlike recent language representation models (Peters et al., 2018a; Radford et al., 2018), BERT is designed to pre-train deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers. As a result, the pre-trained BERT model can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks, such as question answering and language inference, without substantial task-specific architecture modifications. BERT is conceptually simple and empirically powerful. It obtains new state-of-the-art results on eleven natural language processing tasks, including pushing the GLUE score to 80.5 (7.7 point absolute improvement), MultiNLI accuracy to 86.7% (4.6% absolute improvement), SQuAD v1.1 question answering Test F1 to 93.2 (1.5 point absolute improvement) and SQuAD v2.0 Test F1 to 83.1 (5.1 point absolute improvement).

The central building block of convolutional neural networks (CNNs) is the convolution operator, which enables networks to construct informative features by fusing both spatial and channel-wise information within local receptive fields at each layer. A broad range of prior research has investigated the spatial component of this relationship, seeking to strengthen the representational power of a CNN by enhancing the quality of spatial encodings throughout its feature hierarchy. In this work, we focus instead on the channel relationship and propose a novel architectural unit, which we term the “Squeeze-and-Excitation” (SE) block, that adaptively recalibrates channel-wise feature responses by explicitly modelling interdependencies between channels. We show that these blocks can be stacked together to form SENet architectures that generalise extremely effectively across different datasets. We further demonstrate that SE blocks bring significant improvements in performance for existing state-of-the-art CNNs at slight additional computational cost. Squeeze-and-Excitation Networks formed the foundation of our ILSVRC 2017 classification submission which won first place and reduced the top-5 error to 2.251 percent, surpassing the winning entry of 2016 by a relative improvement of ${\sim }$ ∼ 25 percent. Models and code are available at https://github.com/hujie-frank/SENet .

/pdf/squeeze-and-excitation-networks-50icgwqgeu.pdf

Squeeze-and-Excitation Networks

Language model pretraining has led to significant performance gains but careful comparison between different approaches is challenging. Training is computationally expensive, often done on private datasets of different sizes, and, as we will show, hyperparameter choices have significant impact on the final results. We present a replication study of BERT pretraining (Devlin et al., 2019) that carefully measures the impact of many key hyperparameters and training data size. We find that BERT was significantly undertrained, and can match or exceed the performance of every model published after it. Our best model achieves state-of-the-art results on GLUE, RACE and SQuAD. These results highlight the importance of previously overlooked design choices, and raise questions about the source of recently reported improvements. We release our models and code.

/pdf/roberta-a-robustly-optimized-bert-pretraining-approach-2rj0bdyhim.pdf

RoBERTa: A Robustly Optimized BERT Pretraining Approach

While the Transformer architecture has become the de-facto standard for natural language processing tasks, its applications to computer vision remain limited. In vision, attention is either applied in conjunction with convolutional networks, or used to replace certain components of convolutional networks while keeping their overall structure in place. We show that this reliance on CNNs is not necessary and a pure transformer applied directly to sequences of image patches can perform very well on image classification tasks. When pre-trained on large amounts of data and transferred to multiple mid-sized or small image recognition benchmarks (ImageNet, CIFAR-100, VTAB, etc.), Vision Transformer (ViT) attains excellent results compared to state-of-the-art convolutional networks while requiring substantially fewer computational resources to train.

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

Learning object-centric representations of complex scenes is a promising step towards enabling efficient abstract reasoning from low-level perceptual features Yet, most deep learning approaches learn distributed representations that do not capture the compositional properties of natural scenes In this paper, we present the Slot Attention module, an architectural component that interfaces with perceptual representations such as the output of a convolutional neural network and produces a set of task-dependent abstract representations which we call slots These slots are exchangeable and can bind to any object in the input by specializing through a competitive procedure over multiple rounds of attention We empirically demonstrate that Slot Attention can extract object-centric representations that enable generalization to unseen compositions when trained on unsupervised object discovery and supervised property prediction tasks

Object-Centric Learning with Slot Attention

The disclosed subject matter provides a system, computer readable storage medium, and a method providing an audio and textual transcript of a communication A conferencing services may receive audio or audio visual signals from a plurality of different devices that receive voice communications from participants in a communication, such as a chat or teleconference The audio signals representing voice (speech) communications input into respective different devices by the participants A translation services server may receive over a separate communication channel the audio signals for translation into a second language As managed by the translation services server, the audio signals may be converted into textual data The textual data may be translated into text of different languages based the language preferences of the end user devices in the teleconference The translated text may be further translated into audio signals

Auto-translation for multi user audio and video

We propose a general method to watermark and probabilistically identify the structured outputs of machine learning algorithms. Our method is robust to local editing operations and provides well defined trade-offs between the ability to identify algorithm outputs and the quality of the watermarked output. Unlike previous work in the field, our approach does not rely on controlling the inputs to the algorithm and provides probabilistic guarantees on the ability to identify collections of results from one's own algorithm. We present an application in statistical machine translation, where machine translated output is watermarked at minimal loss in translation quality and detected with high recall.

/pdf/watermarking-the-outputs-of-structured-prediction-with-an-3zt3fnfujo.pdf

Watermarking the Outputs of Structured Prediction with an application in Statistical Machine Translation.

Temporal resolution systems are traditionally tuned to a particular language, requiring significant human effort to translate them to new languages. We present a language independent semantic parser for learning the interpretation of temporal phrases given only a corpus of utterances and the times they reference. We make use of a latent parse that encodes a language-flexible representation of time, and extract rich features over both the parse and associated temporal semantics. The parameters of the model are learned using a weakly supervised bootstrapping approach, without the need for manually tuned parameters or any other language expertise. We achieve state-of-the-art accuracy on all languages in the TempEval2 temporal normalization task, reporting a 4% improvement in both English and Spanish accuracy, and to our knowledge the first results for four other languages.

/pdf/language-independent-discriminative-parsing-of-temporal-23mgsp4o99.pdf

Language-Independent Discriminative Parsing of Temporal Expressions

The present disclosure describes a teleconferencing system that may use a virtual participant processor to translate language content of the teleconference into each participant's spoken language without additional user inputs. The virtual participant processor may connect to the teleconference as do the other participants. The virtual participant processor may intercept all text or audio data that was previously exchanged between the participants may now be intercepted by the virtual participant processor. Upon obtaining a partial or complete language recognition result or making a language preference determination, the virtual participant processor may call a translation engine appropriate for each of the participants. The virtual participant processor may send the resulting translation to a teleconference management processor. The teleconference management processor may deliver the respective translated text or audio data to the appropriate participant.

Jakob Uszkoreit

Papers

Object-Centric Learning with Slot Attention

Auto-translation for multi user audio and video

Watermarking the Outputs of Structured Prediction with an application in Statistical Machine Translation.

Language-Independent Discriminative Parsing of Temporal Expressions

Virtual participant-based real-time translation and transcription system for audio and video teleconferences