The development of neural networks and pretraining techniques has spawned many sentence-level tagging systems that achieved superior performance on typical benchmarks. However, a relatively less discussed topic is what if more context information is introduced into current top-scoring tagging systems. Although several existing works have attempted to shift tagging systems from sentence-level to document-level, there is still no consensus conclusion about when and why it works, which limits the applicability of the larger-context approach in tagging tasks. In this paper, instead of pursuing a state-of-the-art tagging system by architectural exploration, we focus on investigating when and why the larger-context training, as a general strategy, can work. To this end, we conduct a thorough comparative study on four proposed aggregators for context information collecting and present an attribute-aided evaluation method to interpret the improvement brought by larger-context training. Experimentally, we set up a testbed based on four tagging tasks and thirteen datasets. Hopefully, our preliminary observations can deepen the understanding of larger-context training and enlighten more follow-up works on the use of contextual information.
相關內容
When and Why does a Model Fail? A Human-in-the-loop Error Detection Framework for Sentiment Analysis
Although deep neural networks have been widely employed and proven effective in sentiment analysis tasks, it remains challenging for model developers to assess their models for erroneous predictions that might exist prior to deployment. Once deployed, emergent errors can be hard to identify in prediction run-time and impossible to trace back to their sources. To address such gaps, in this paper we propose an error detection framework for sentiment analysis based on explainable features. We perform global-level feature validation with human-in-the-loop assessment, followed by an integration of global and local-level feature contribution analysis. Experimental results show that, given limited human-in-the-loop intervention, our method is able to identify erroneous model predictions on unseen data with high precision.
Crowdsourcing is widely used to create data for common natural language understanding tasks. Despite the importance of these datasets for measuring and refining model understanding of language, there has been little focus on the crowdsourcing methods used for collecting the datasets. In this paper, we compare the efficacy of interventions that have been proposed in prior work as ways of improving data quality. We use multiple-choice question answering as a testbed and run a randomized trial by assigning crowdworkers to write questions under one of four different data collection protocols. We find that asking workers to write explanations for their examples is an ineffective stand-alone strategy for boosting NLU example difficulty. However, we find that training crowdworkers, and then using an iterative process of collecting data, sending feedback, and qualifying workers based on expert judgments is an effective means of collecting challenging data. But using crowdsourced, instead of expert judgments, to qualify workers and send feedback does not prove to be effective. We observe that the data from the iterative protocol with expert assessments is more challenging by several measures. Notably, the human--model gap on the unanimous agreement portion of this data is, on average, twice as large as the gap for the baseline protocol data.
A code completion system suggests future code elements to developers given a partially-complete code snippet. Code completion is one of the most useful features in Integrated Development Environments (IDEs). Currently, most code completion techniques predict a single token at a time. In this paper, we take a further step and discuss the probability of directly completing a whole line of code instead of a single token. We believe suggesting longer code sequences can further improve the efficiency of developers. Recently neural language models have been adopted as a preferred approach for code completion, and we believe these models can still be applied to full-line code completion with a few improvements. We conduct our experiments on two real-world python corpora and evaluate existing neural models based on source code tokens or syntactical actions. The results show that neural language models can achieve acceptable results on our tasks, with significant room for improvements.
To make deliberate progress towards more intelligent and more human-like artificial systems, we need to be following an appropriate feedback signal: we need to be able to define and evaluate intelligence in a way that enables comparisons between two systems, as well as comparisons with humans. Over the past hundred years, there has been an abundance of attempts to define and measure intelligence, across both the fields of psychology and AI. We summarize and critically assess these definitions and evaluation approaches, while making apparent the two historical conceptions of intelligence that have implicitly guided them. We note that in practice, the contemporary AI community still gravitates towards benchmarking intelligence by comparing the skill exhibited by AIs and humans at specific tasks such as board games and video games. We argue that solely measuring skill at any given task falls short of measuring intelligence, because skill is heavily modulated by prior knowledge and experience: unlimited priors or unlimited training data allow experimenters to "buy" arbitrary levels of skills for a system, in a way that masks the system's own generalization power. We then articulate a new formal definition of intelligence based on Algorithmic Information Theory, describing intelligence as skill-acquisition efficiency and highlighting the concepts of scope, generalization difficulty, priors, and experience. Using this definition, we propose a set of guidelines for what a general AI benchmark should look like. Finally, we present a benchmark closely following these guidelines, the Abstraction and Reasoning Corpus (ARC), built upon an explicit set of priors designed to be as close as possible to innate human priors. We argue that ARC can be used to measure a human-like form of general fluid intelligence and that it enables fair general intelligence comparisons between AI systems and humans.
In this paper, we investigate the challenges of using reinforcement learning agents for question-answering over knowledge graphs for real-world applications. We examine the performance metrics used by state-of-the-art systems and determine that they are inadequate for such settings. More specifically, they do not evaluate the systems correctly for situations when there is no answer available and thus agents optimized for these metrics are poor at modeling confidence. We introduce a simple new performance metric for evaluating question-answering agents that is more representative of practical usage conditions, and optimize for this metric by extending the binary reward structure used in prior work to a ternary reward structure which also rewards an agent for not answering a question rather than giving an incorrect answer. We show that this can drastically improve the precision of answered questions while only not answering a limited number of previously correctly answered questions. Employing a supervised learning strategy using depth-first-search paths to bootstrap the reinforcement learning algorithm further improves performance.
In this work, we present novel methods to adapt visual QA models for community QA tasks of practical significance - automated question category classification and finding experts for question answering - on questions containing both text and image. To the best of our knowledge, this is the first work to tackle the multimodality challenge in CQA, and is an enabling step towards basic question-answering on image-based CQA. First, we analyze the differences between visual QA and community QA datasets, discussing the limitations of applying VQA models directly to CQA tasks, and then we propose novel augmentations to VQA-based models to best address those limitations. Our model, with the augmentations of an image-text combination method tailored for CQA and use of auxiliary tasks for learning better grounding features, significantly outperforms the text-only and VQA model baselines for both tasks on real-world CQA data from Yahoo! Chiebukuro, a Japanese counterpart of Yahoo! Answers.
Many natural language processing tasks solely rely on sparse dependencies between a few tokens in a sentence. Soft attention mechanisms show promising performance in modeling local/global dependencies by soft probabilities between every two tokens, but they are not effective and efficient when applied to long sentences. By contrast, hard attention mechanisms directly select a subset of tokens but are difficult and inefficient to train due to their combinatorial nature. In this paper, we integrate both soft and hard attention into one context fusion model, "reinforced self-attention (ReSA)", for the mutual benefit of each other. In ReSA, a hard attention trims a sequence for a soft self-attention to process, while the soft attention feeds reward signals back to facilitate the training of the hard one. For this purpose, we develop a novel hard attention called "reinforced sequence sampling (RSS)", selecting tokens in parallel and trained via policy gradient. Using two RSS modules, ReSA efficiently extracts the sparse dependencies between each pair of selected tokens. We finally propose an RNN/CNN-free sentence-encoding model, "reinforced self-attention network (ReSAN)", solely based on ReSA. It achieves state-of-the-art performance on both Stanford Natural Language Inference (SNLI) and Sentences Involving Compositional Knowledge (SICK) datasets.
With the emergence of Web 2.0, tag recommenders have become important tools, which aim to support users in finding descriptive tags for their bookmarked resources. Although current algorithms provide good results in terms of tag prediction accuracy, they are often designed in a data-driven way and thus, lack a thorough understanding of the cognitive processes that play a role when people assign tags to resources. This thesis aims at modeling these cognitive dynamics in social tagging in order to improve tag recommendations and to better understand the underlying processes. As a first attempt in this direction, we have implemented an interplay between individual micro-level (e.g., categorizing resources or temporal dynamics) and collective macro-level (e.g., imitating other users' tags) processes in the form of a novel tag recommender algorithm. The preliminary results for datasets gathered from BibSonomy, CiteULike and Delicious show that our proposed approach can outperform current state-of-the-art algorithms, such as Collaborative Filtering, FolkRank or Pairwise Interaction Tensor Factorization. We conclude that recommender systems can be improved by incorporating related principles of human cognition.
Movie recommendation systems provide users with ranked lists of movies based on individual's preferences and constraints. Two types of models are commonly used to generate ranking results: long-term models and session-based models. While long-term models represent the interactions between users and movies that are supposed to change slowly across time, session-based models encode the information of users' interests and changing dynamics of movies' attributes in short terms. In this paper, we propose an LSIC model, leveraging Long and Short-term Information in Content-aware movie recommendation using adversarial training. In the adversarial process, we train a generator as an agent of reinforcement learning which recommends the next movie to a user sequentially. We also train a discriminator which attempts to distinguish the generated list of movies from the real records. The poster information of movies is integrated to further improve the performance of movie recommendation, which is specifically essential when few ratings are available. The experiments demonstrate that the proposed model has robust superiority over competitors and sets the state-of-the-art. We will release the source code of this work after publication.
Novel neural models have been proposed in recent years for learning under domain shift. Most models, however, only evaluate on a single task, on proprietary datasets, or compare to weak baselines, which makes comparison of models difficult. In this paper, we re-evaluate classic general-purpose bootstrapping approaches in the context of neural networks under domain shifts vs. recent neural approaches and propose a novel multi-task tri-training method that reduces the time and space complexity of classic tri-training. Extensive experiments on two benchmarks are negative: while our novel method establishes a new state-of-the-art for sentiment analysis, it does not fare consistently the best. More importantly, we arrive at the somewhat surprising conclusion that classic tri-training, with some additions, outperforms the state of the art. We conclude that classic approaches constitute an important and strong baseline.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191