Papers.

Just-in-time performance without warm-up

Scala has been developed as a language that deeply integrates with the Java ecosystem. It offers seamless interoperability with existing Java libraries. Since the Scala compiler targets Java bytecode, Scala programs have access to high-performance runtimes including the HotSpot virtual machine. HotSpot provides impressive performance results achieved via just-in-time compilation. It starts program execution in interpreter mode, collecting profile feedback about called methods. This information allows HotSpot to identify hot spots in the program, which are then compiled on the fly to native code. This compilation scheme enables high peak performance at the cost of warmup time required to collect the profile data and perform just-in-time compilation. This is a good example of the traditional tradeoff between ahead-of-time (AOT) and just-in-time (JIT) compilation. With AOT, compilers have less information, but the runtime story is reasonably straightforward. With JIT, compilers have more information, which enables advanced optimizations, but the runtime story becomes complicated. In this dissertation, we present the design and implementation of Scala Native, an optimizing compiler for Scala. With Scala Native, Scala programs are compiled ahead of time, which avoids runtime compilation and enables instant startup times. On the other hand, Scala Native is able to match and supersede the peak performance of HotSpot on our benchmarks. In addition to that, Scala Native is a general-purpose Scala compiler - programs compiled by Scala Native closely match the behavior of programs compiled by the Scala compiler. First, we introduce NIR, an intermediate representation designed with ahead-of-time compilation in mind. NIR represents programs in the single-static assignment form and has support for object-oriented features such as virtual dispatch and multiple inheritance. This representation is a key enabler of our compilation and optimization pipeline. Secondly, we present Inter

The many faces of resilience

fastai: A Layered API for Deep Learning

fastai is a deep learning library which provides practitioners with high-level components that can quickly and easily provide state-of-the-art results in standard deep learning domains, and provides researchers with low-level components that can be mixed and matched to build new approaches. It aims to do both things without substantial compromises in ease of use, flexibility, or performance. This is possible thanks to a carefully layered architecture, which expresses common underlying patterns of many deep learning and data processing techniques in terms of decoupled abstractions. These abstractions can be expressed concisely and clearly by leveraging the dynamism of the underlying Python language and the flexibility of the PyTorch library. fastai includes: a new type dispatch system for Python along with a semantic type hierarchy for tensors; a GPU-optimized computer vision library which can be extended in pure Python; an optimizer which refactors out the common functionality of modern optimizers into two basic pieces, allowing optimization algorithms to be implemented in 4–5 lines of code; a novel 2-way callback system that can access any part of the data, model, or optimizer and change it at any point during training; a new data block API; and much more. We used this library to successfully create a complete deep learning course, which we were able to write more quickly than using previous approaches, and the code was more clear. The library is already in wide use in research, industry, and teaching.

A New Acoustic-Based Pronunciation Distance Measure

Η Τεχνητή Νοημοσύνη συναντά την Αυτόματη Αναγνώριση Ζωγραφιάς

Exploration Based Language Learning for Text-Based Games

This work presents an exploration and imitation-learning-based agent capable of state-of-the-art performance in playing text-based computer games. These games are of interest as they can be seen as a testbed for language understanding, problem-solving, and language generation by artificial agents. Moreover, they provide a learning setting in which these skills can be acquired through interactions with an environment rather than using fixed corpora. One aspect that makes these games particularly challenging for learning agents is the combinatorially large action space. Existing methods for solving text-based games are limited to games that are either very simple or have an action space restricted to a predetermined set of admissible actions. In this work, we propose to use the exploration approach of Go-Explore for solving text-based games. More specifically, in an initial exploration phase, we first extract trajectories with high rewards, after which we train a policy to solve the game by imitating these trajectories. Our experiments show that this approach outperforms existing solutions in solving text-based games, and it is more sample efficient in terms of the number of interactions with the environment. Moreover, we show that the learned policy can generalize better than existing solutions to unseen games without using any restriction on the action space.

Special Issue "On Defining Artificial Intelligence" - Commentaries and Author's Response

A collection of commentaries on the definition of artificial intelligence, with the authors' response; Chollet contributes a commentary consistent with his 'On the Measure of Intelligence' position.

Characterization of Anorexia Nervosa on Social Media: Textual, Visual, Relational, Behavioral, and Demographical Analysis (Preprint)

Detection of Suicidal Ideation on Social Media: Multimodal, Relational, and Behavioral Analysis (Preprint)

Detection of Suicidal Ideation on Social Media: Multimodal, Relational, and Behavioral Analysis

Punctuation Prediction in Spontaneous Conversations: Can We Mitigate ASR Errors with Retrofitted Word Embeddings?

Automatic Speech Recognition (ASR) systems introduce word errors, which often confuse punctuation prediction models, turning punctuation restoration into a challenging task. These errors usually take the form of homonyms. We show how retrofitting of the word embeddings on the domain-specific data can mitigate ASR errors. Our main contribution is a method for better alignment of homonym embeddings and the validation of the presented method on the punctuation prediction task. We record the absolute improvement in punctuation prediction accuracy between 6.2% (for question marks) to 9% (for periods) when compared with the state-of-the-art model.

Limiting Bias from Test-Control Interference in Online Marketplace Experiments

UniK-QA: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering

We study open-domain question answering with structured, unstructured and semi-structured knowledge sources, including text, tables, lists and knowledge bases. Departing from prior work, we propose a unifying approach that homogenizes all sources by reducing them to text and applies the retriever-reader model which has so far been limited to text sources only. Our approach greatly improves the results on knowledge-base QA tasks by 11 points, compared to latest graph-based methods. More importantly, we demonstrate that our unified knowledge (UniK-QA) model is a simple and yet effective way to combine heterogeneous sources of knowledge, advancing the state-of-the-art results on two popular question answering benchmarks, NaturalQuestions and WebQuestions, by 3.5 and 2.6 points, respectively. The code of UniK-QA is available at: https://github.com/facebookresearch/UniK-QA.

Technical Question Answering across Tasks and Domains

Building automatic technical support system is an important yet challenge task. Conceptually, to answer a user question on a technical forum, a human expert has to first retrieve relevant documents, and then read them carefully to identify the answer snippet. Despite huge success the researchers have achieved in coping with general domain question answering (QA), much less attentions have been paid for investigating technical QA. Specifically, existing methods suffer from several unique challenges (i) the question and answer rarely overlaps substantially and (ii) very limited data size. In this paper, we propose a novel framework of deep transfer learning to effectively address technical QA across tasks and domains. To this end, we present an adjustable joint learning approach for document retrieval and reading comprehension tasks. Our experiments on the TechQA demonstrates superior performance compared with state-of-the-art methods.

MTOP: A Comprehensive Multilingual Task-Oriented Semantic Parsing Benchmark

Scaling semantic parsing models for task-oriented dialog systems to new languages is often expensive and time-consuming due to the lack of available datasets. Available datasets suffer from several shortcomings: a) they contain few languages b) they contain small amounts of labeled examples per language c) they are based on the simple intent and slot detection paradigm for non-compositional queries. In this paper, we present a new multilingual dataset, called MTOP, comprising of 100k annotated utterances in 6 languages across 11 domains. We use this dataset and other publicly available datasets to conduct a comprehensive benchmarking study on using various state-of-the-art multilingual pre-trained models for task-oriented semantic parsing. We achieve an average improvement of +6.3 points on Slot F1 for the two existing multilingual datasets, over best results reported in their experiments. Furthermore, we demonstrate strong zero-shot performance using pre-trained models combined with automatic translation and alignment, and a proposed distant supervision method to reduce the noise in slot label projection.

Intrinsic Dimensionality Explains the Effectiveness of Language Model Fine-Tuning

Although pretrained language models can be fine-tuned to produce state-of-the-art results for a very wide range of language understanding tasks, the dynamics of this process are not well understood, especially in the low data regime. Why can we use relatively vanilla gradient descent algorithms (e.g., without strong regularization) to tune a model with hundreds of millions of parameters on datasets with only hundreds or thousands of labeled examples? In this paper, we argue that analyzing fine-tuning through the lens of intrinsic dimension provides us with empirical and theoretical intuitions to explain this remarkable phenomenon. We empirically show that common pre-trained models have a very low intrinsic dimension; in other words, there exists a low dimension reparameterization that is as effective for fine-tuning as the full parameter space. For example, by optimizing only 200 trainable parameters randomly projected back into the full space, we can tune a RoBERTa model to achieve 90% of the full parameter performance levels on MRPC. Furthermore, we empirically show that pre-training implicitly minimizes intrinsic dimension and, perhaps surprisingly, larger models tend to have lower intrinsic dimension after a fixed number of pre-training updates, at least in part explaining their extreme effectiveness. Lastly, we connect intrinsic dimensionality with low dimensional task representations and compression based generalization bounds to provide intrinsic-dimension-based generalization bounds that are independent of the full parameter count.

Beyond Domain APIs: Task-oriented Conversational Modeling with Unstructured Knowledge Access

A Multilingual Neural Machine Translation Model for Biomedical Data

COVIDLies: Detecting COVID-19 Misinformation on Social Media

The ongoing pandemic has heightened the need for developing tools to flag COVID-19-related misinformation on the internet, specifically on social media such as Twitter. However, due to novel language and the rapid change of information, existing misinformation detection datasets are not effective for evaluating systems designed to detect misin-formation on this topic. Misinformation detection can be divided into two sub-tasks: (i) retrieval of misconceptions relevant to posts being checked for veracity, and (ii) stance detection to identify whether the posts Agree , Dis-agree , or express No Stance towards the retrieved misconceptions. To facilitate research on this task, we release C OVID L IES 1 , a dataset of 6761 expert-annotated tweets to evaluate the performance of misinformation detection systems on 86 different pieces of COVID-19 related misinformation. We evaluate existing NLP systems on this dataset, providing initial benchmarks and identifying key challenges for future models to improve upon.

Policy-Driven Neural Response Generation for Knowledge-Grounded Dialog Systems

Controllable Text Generation with Focused Variation

This work introduces Focused-Variation Network (FVN), a novel model to control language generation. The main problems in previous controlled language generation models range from the difficulty of generating text according to the given attributes, to the lack of diversity of the generated texts. FVN addresses these issues by learning disjoint discrete latent spaces for each attribute inside codebooks, which allows for both controllability and diversity, while at the same time generating fluent text. We evaluate FVN on two text generation datasets with annotated content and style, and show state-of-the-art performance as assessed by automatic and human evaluations.

WER we are and WER we think we are

Natural language processing of conversational speech requires the availability of high-quality transcripts. In this paper, we express our skepticism towards the recent reports of very low Word Error Rates (WERs) achieved by modern Automatic Speech Recognition (ASR) systems on benchmark datasets. We outline several problems with popular benchmarks and compare three state-of-the-art commercial ASR systems on an internal dataset of real-life spontaneous human conversations and HUB’05 public benchmark. We show that WERs are significantly higher than the best reported results. We formulate a set of guidelines which may aid in the creation of real-life, multi-domain datasets with high quality annotations for training and testing of robust ASR systems.

Evaluating Models’ Local Decision Boundaries via Contrast Sets

Standard test sets for supervised learning evaluate in-distribution generalization. Unfortunately, when a dataset has systematic gaps (e.g., annotation artifacts), these evaluations are misleading: a model can learn simple decision rules that perform well on the test set but do not capture the abilities a dataset is intended to test. We propose a more rigorous annotation paradigm for NLP that helps to close systematic gaps in the test data. In particular, after a dataset is constructed, we recommend that the dataset authors manually perturb the test instances in small but meaningful ways that (typically) change the gold label, creating contrast sets. Contrast sets provide a local view of a model’s decision boundary, which can be used to more accurately evaluate a model’s true linguistic capabilities. We demonstrate the efficacy of contrast sets by creating them for 10 diverse NLP datasets (e.g., DROP reading comprehension, UD parsing, and IMDb sentiment analysis). Although our contrast sets are not explicitly adversarial, model performance is significantly lower on them than on the original test sets—up to 25% in some cases. We release our contrast sets as new evaluation benchmarks and encourage future dataset construction efforts to follow similar annotation processes.

Sound Natural: Content Rephrasing in Dialog Systems

We introduce a new task of rephrasing for a more natural virtual assistant. Currently, virtual assistants work in the paradigm of intent slot tagging and the slot values are directly passed as-is to the execution engine. However, this setup fails in some scenarios such as messaging when the query given by the user needs to be changed before repeating it or sending it to another user. For example, for queries like 'ask my wife if she can pick up the kids' or 'remind me to take my pills', we need to rephrase the content to 'can you pick up the kids' and 'take your pills' In this paper, we study the problem of rephrasing with messaging as a use case and release a dataset of 3000 pairs of original query and rephrased query. We show that BART, a pre-trained transformers-based masked language model with auto-regressive decoding, is a strong baseline for the task, and show improvements by adding a copy-pointer and copy loss to it. We analyze different tradeoffs of BART-based and LSTM-based seq2seq models, and propose a distilled LSTM-based seq2seq as the best practical model.

Low-Resource Domain Adaptation for Compositional Task-Oriented Semantic Parsing

Task-oriented semantic parsing is a critical component of virtual assistants, which is responsible for understanding the user's intents (set reminder, play music, etc.). Recent advances in deep learning have enabled several approaches to successfully parse more complex queries (Gupta et al., 2018; Rongali et al.,2020), but these models require a large amount of annotated training data to parse queries on new domains (e.g. reminder, music). In this paper, we focus on adapting task-oriented semantic parsers to low-resource domains, and propose a novel method that outperforms a supervised neural model at a 10-fold data reduction. In particular, we identify two fundamental factors for low-resource domain adaptation: better representation learning and better training techniques. Our representation learning uses BART (Lewis et al., 2019) to initialize our model which outperforms encoder-only pre-trained representations used in previous work. Furthermore, we train with optimization-based meta-learning (Finn et al., 2017) to improve generalization to low-resource domains. This approach significantly outperforms all baseline methods in the experiments on a newly collected multi-domain task-oriented semantic parsing dataset (TOPv2), which we release to the public.

Conversational Semantic Parsing

The structured representation for semantic parsing in task-oriented assistant systems is geared towards simple understanding of one-turn queries. Due to the limitations of the representation, the session-based properties such as co-reference resolution and context carryover are processed downstream in a pipelined system. In this paper, we propose a semantic representation for such task-oriented conversational systems that can represent concepts such as co-reference and context carryover, enabling comprehensive understanding of queries in a session. We release a new session-based, compositional task-oriented parsing dataset of 20k sessions consisting of 60k utterances. Unlike Dialog State Tracking Challenges, the queries in the dataset have compositional forms. We propose a new family of Seq2Seq models for the session-based parsing above, which achieve better or comparable performance to the current state-of-the-art on ATIS, SNIPS, TOP and DSTC2. Notably, we improve the best known results on DSTC2 by up to 5 points for slot-carryover.

Eliciting Knowledge from Language Models Using Automatically Generated Prompts

The remarkable success of pretrained language models has motivated the study of what kinds of knowledge these models learn during pretraining. Reformulating tasks as fill-in-the-blanks problems (e.g., cloze tests) is a natural approach for gauging such knowledge, however, its usage is limited by the manual effort and guesswork required to write suitable prompts. To address this, we develop AutoPrompt, an automated method to create prompts for a diverse set of tasks, based on a gradient-guided search. Using AutoPrompt, we show that masked language models (MLMs) have an inherent capability to perform sentiment analysis and natural language inference without additional parameters or finetuning, sometimes achieving performance on par with recent state-of-the-art supervised models. We also show that our prompts elicit more accurate factual knowledge from MLMs than the manually created prompts on the LAMA benchmark, and that MLMs can be used as relation extractors more effectively than supervised relation extraction models. These results demonstrate that automatically generated prompts are a viable parameter-free alternative to existing probing methods, and as pretrained LMs become more sophisticated and capable, potentially a replacement for finetuning.

How to Make Neural Natural Language Generation as Reliable as Templates in Task-Oriented Dialogue

Detecting Independent Pronoun Bias with Partially-Synthetic Data Generation

HuggingFace's Transformers: State-of-the-art Natural Language Processing

Identifying Incorrect Labels in the CoNLL-2003 Corpus

The CoNLL-2003 corpus for English-language named entity recognition (NER) is one of the most influential corpora for NER model research. A large number of publications, including many landmark works, have used this corpus as a source of ground truth for NER tasks. In this paper, we examine this corpus and identify over 1300 incorrect labels (out of 35089 in the corpus). In particular, the number of incorrect labels in the test fold is comparable to the number of errors that state-of-the-art models make when running inference over this corpus. We describe the process by which we identified these incorrect labels, using novel variants of techniques from semi-supervised learning. We also summarize the types of errors that we found, and we revisit several recent results in NER in light of the corrected data. Finally, we show experimentally that our corrections to the corpus have a positive impact on three state-of-the-art models.

Joint Modeling of Arguments for Event Understanding

We recognize the task of event argument linking in documents as similar to that of intent slot resolution in dialogue, providing a Transformer-based model that extends from a recently proposed solution to resolve references to slots. The approach allows for joint consideration of argument candidates given a detected event, which we illustrate leads to state-of-the-art performance in multi-sentence argument linking.

Hierarchical Entity Typing via Multi-level Learning to Rank

We propose a novel method for hierarchical entity classification that embraces ontological structure at both training and during prediction. At training, our novel multi-level learning-to-rank loss compares positive types against negative siblings according to the type tree. During prediction, we define a coarse-to-fine decoder that restricts viable candidates at each level of the ontology based on already predicted parent type(s). Our approach significantly outperform prior work on strict accuracy, demonstrating the effectiveness of our method.

Shape of Synth to Come: Why We Should Use Synthetic Data for English Surface Realization

Beyond Accuracy: Behavioral Testing of NLP Models with CheckList

Although measuring held-out accuracy has been the primary approach to evaluate generalization, it often overestimates the performance of NLP models, while alternative approaches for evaluating models either focus on individual tasks or on specific behaviors. Inspired by principles of behavioral testing in software engineering, we introduce CheckList, a task-agnostic methodology for testing NLP models. CheckList includes a matrix of general linguistic capabilities and test types that facilitate comprehensive test ideation, as well as a software tool to generate a large and diverse number of test cases quickly. We illustrate the utility of CheckList with tests for three tasks, identifying critical failures in both commercial and state-of-art models. In a user study, a team responsible for a commercial sentiment analysis model found new and actionable bugs in an extensively tested model. In another user study, NLP practitioners with CheckList created twice as many tests, and found almost three times as many bugs as users without it.

Recommendations to Handle Health-related Small Imbalanced Data in Machine Learning

Is Data Privacy The Price We Must Pay to Survive a Pandemic?

Human Language Technology

Improved Consistency Regularization for GANs

Recent work has increased the performance of Generative Adversarial Networks (GANs) by enforcing a consistency cost on the discriminator. We improve on this technique in several ways. We first show that consistency regularization can introduce artifacts into the GAN samples and explain how to fix this issue. We then propose several modifications to the consistency regularization procedure designed to improve its performance. We carry out extensive experiments quantifying the benefit of our improvements. For unconditional image synthesis on CIFAR-10 and CelebA, our modifications yield the best known FID scores on various GAN architectures. For conditional image synthesis on CIFAR-10, we improve the state-of-the-art FID score from 11.48 to 9.21. Finally, on ImageNet-2012, we apply our technique to the original BigGAN model and improve the FID from 6.66 to 5.38, which is the best score at that model size.

Just Ask: An Interactive Learning Framework for Vision and Language Navigation

How to Handle Health-Related Small Imbalanced Data in Machine Learning?

CONTENTS

Frontmatter

Introduction

Acknowledgements

List of Illustrations

Intercultural Screen Adaptation

Jointly optimizing preprocessing and inference for DNN-based visual analytics

Winds from seattle

Delta Lake: High-Performance ACID Table Storage over Cloud Object Stores

Cloud object stores such as Amazon S3 are some of the largest and most cost-effective storage systems on the planet, making them an attractive target to store large data warehouses and data lakes. Unfortunately, their implementation as key-value stores makes it difficult to achieve ACID transactions and high performance: metadata operations such as listing objects are expensive, and consistency guarantees are limited. In this paper, we present Delta Lake, an open source ACID table storage layer over cloud object stores initially developed at Databricks. Delta Lake uses a transaction log that is compacted into Apache Parquet format to provide ACID properties, time travel, and significantly faster metadata operations for large tabular datasets (e.g., the ability to quickly search billions of table partitions for those relevant to a query). It also leverages this design to provide high-level features such as automatic data layout optimization, upserts, caching, and audit logs. Delta Lake tables can be accessed from Apache Spark, Hive, Presto, Redshift and other systems. Delta Lake is deployed at thousands of Databricks customers that process exabytes of data per day, with the largest instances managing exabyte-scale datasets and billions of objects.