J. D. Williams

Cecilia Aas, Hisham Abdelsalam, Irina Belousova et al.

It has recently become feasible to run personal digital assistants on phones and other personal devices. In this paper we describe a design for a natural language understanding system that runs on device. In comparison to a server-based assistant, this system is more private, more reliable, faster, more expressive, and more accurate. We describe what led to key choices about architecture and technologies. For example, some approaches in the dialog systems literature are difficult to maintain over time in a deployment setting. We hope that sharing learnings from our practical experiences may help inform future work in the research community.

David Q. Sun, Artem Abzaliev, Hadas Kotek et al.

Controversy is a reflection of our zeitgeist, and an important aspect to any discourse. The rise of large language models (LLMs) as conversational systems has increased public reliance on these systems for answers to their various questions. Consequently, it is crucial to systematically examine how these models respond to questions that pertaining to ongoing debates. However, few such datasets exist in providing human-annotated labels reflecting the contemporary discussions. To foster research in this area, we propose a novel construction of a controversial questions dataset, expanding upon the publicly released Quora Question Pairs Dataset. This dataset presents challenges concerning knowledge recency, safety, fairness, and bias. We evaluate different LLMs using a subset of this dataset, illuminating how they handle controversial issues and the stances they adopt. This research ultimately contributes to our understanding of LLMs' interaction with controversial issues, paving the way for improvements in their comprehension and handling of complex societal debates.

Zidi Xiu, Kai-Chen Cheng, David Q. Sun et al.

With the growing popularity of intelligent assistants (IAs), evaluating IA quality becomes an increasingly active field of research. This paper identifies and quantifies the feedback effect, a novel component in IA-user interactions: how the capabilities and limitations of the IA influence user behavior over time. First, we demonstrate that unhelpful responses from the IA cause users to delay or reduce subsequent interactions in the short term via an observational study. Next, we expand the time horizon to examine behavior changes and show that as users discover the limitations of the IA's understanding and functional capabilities, they learn to adjust the scope and wording of their requests to increase the likelihood of receiving a helpful response from the IA. Our findings highlight the impact of the feedback effect at both the micro and meso levels. We further discuss its macro-level consequences: unsatisfactory interactions continuously reduce the likelihood and diversity of future user engagements in a feedback loop.

Deepak Muralidharan, Joel Ruben Antony Moniz, Weicheng Zhang et al.

Named entity recognition (NER) is usually developed and tested on text from well-written sources. However, in intelligent voice assistants, where NER is an important component, input to NER may be noisy because of user or speech recognition error. In applications, entity labels may change frequently, and non-textual properties like topicality or popularity may be needed to choose among alternatives. We describe a NER system intended to address these problems. We test and train this system on a proprietary user-derived dataset. We compare with a baseline text-only NER system; the baseline enhanced with external gazetteers; and the baseline enhanced with the search and indirect labelling techniques we describe below. The final configuration gives around 6% reduction in NER error rate. We also show that this technique improves related tasks, such as semantic parsing, with an improvement of up to 5% in error rate.

David Q. Sun, Hadas Kotek, Christopher Klein et al.

This paper develops and implements a scalable methodology for (a) estimating the noisiness of labels produced by a typical crowdsourcing semantic annotation task, and (b) reducing the resulting error of the labeling process by as much as 20-30% in comparison to other common labeling strategies. Importantly, this new approach to the labeling process, which we name Dynamic Automatic Conflict Resolution (DACR), does not require a ground truth dataset and is instead based on inter-project annotation inconsistencies. This makes DACR not only more accurate but also available to a broad range of labeling tasks. In what follows we present results from a text classification task performed at scale for a commercial personal assistant, and evaluate the inherent ambiguity uncovered by this annotation strategy as compared to other common labeling strategies.

Alkesh Patel, Akanksha Bindal, Hadas Kotek et al.

Generating natural, diverse, and meaningful questions from images is an essential task for multimodal assistants as it confirms whether they have understood the object and scene in the images properly. The research in visual question answering (VQA) and visual question generation (VQG) is a great step. However, this research does not capture questions that a visually-abled person would ask multimodal assistants. Recently published datasets such as KB-VQA, FVQA, and OK-VQA try to collect questions that look for external knowledge which makes them appropriate for multimodal assistants. However, they still contain many obvious and common-sense questions that humans would not usually ask a digital assistant. In this paper, we provide a new benchmark dataset that contains questions generated by human annotators keeping in mind what they would ask multimodal digital assistants. Large scale annotations for several hundred thousand images are expensive and time-consuming, so we also present an effective way of automatically generating questions from unseen images. In this paper, we present an approach for generating diverse and meaningful questions that consider image content and metadata of image (e.g., location, associated keyword). We evaluate our approach using standard evaluation metrics such as BLEU, METEOR, ROUGE, and CIDEr to show the relevance of generated questions with human-provided questions. We also measure the diversity of generated questions using generative strength and inventiveness metrics. We report new state-of-the-art results on the public and our datasets.

Jianpeng Cheng, Devang Agrawal, Hector Martinez Alonso et al.

We consider a new perspective on dialog state tracking (DST), the task of estimating a user's goal through the course of a dialog. By formulating DST as a semantic parsing task over hierarchical representations, we can incorporate semantic compositionality, cross-domain knowledge sharing and co-reference. We present TreeDST, a dataset of 27k conversations annotated with tree-structured dialog states and system acts. We describe an encoder-decoder framework for DST with hierarchical representations, which leads to 20% improvement over state-of-the-art DST approaches that operate on a flat meaning space of slot-value pairs.

Deepak Muralidharan, Justine Kao, Xiao Yang et al.

Personal assistant AI systems such as Siri, Cortana, and Alexa have become widely used as a means to accomplish tasks through natural language commands. However, components in these systems generally rely on supervised machine learning algorithms that require large amounts of hand-annotated training data, which is expensive and time consuming to collect. The ability to incorporate unsupervised, weakly supervised, or distantly supervised data holds significant promise in overcoming this bottleneck. In this paper, we describe a framework that leverages user engagement signals (user behaviors that demonstrate a positive or negative response to content) to automatically create granular entity labels for training data augmentation. Strategies such as multi-task learning and validation using an external knowledge base are employed to incorporate the engagement annotated data and to boost the model's accuracy on a sequence labeling task. Our results show that learning from data automatically labeled by user engagement signals achieves significant accuracy gains in a production deep learning system, when measured on both the sequence labeling task as well as on user facing results produced by the system end-to-end. We believe this is the first use of user engagement signals to help generate training data for a sequence labeling task on a large scale, and can be applied in practical settings to speed up new feature deployment when little human annotated data is available.

Xi C. Chen, Adithya Sagar, Justine T. Kao et al.

We describe a method for selecting relevant new training data for the LSTM-based domain selection component of our personal assistant system. Adding more annotated training data for any ML system typically improves accuracy, but only if it provides examples not already adequately covered in the existing data. However, obtaining, selecting, and labeling relevant data is expensive. This work presents a simple technique that automatically identifies new helpful examples suitable for human annotation. Our experimental results show that the proposed method, compared with random-selection and entropy-based methods, leads to higher accuracy improvements given a fixed annotation budget. Although developed and tested in the setting of a commercial intelligent assistant, the technique is of wider applicability.

Matthew Hausknecht, Ricky Loynd, Greg Yang et al.

Interactive Fiction (IF) games are complex textual decision making problems. This paper introduces NAIL, an autonomous agent for general parser-based IF games. NAIL won the 2018 Text Adventure AI Competition, where it was evaluated on twenty unseen games. This paper describes the architecture, development, and insights underpinning NAIL's performance.

Hai Wang, Jason D. Williams, SingBing Kang

We show how we can globally edit images using textual instructions: given a source image and a textual instruction for the edit, generate a new image transformed under this instruction. To tackle this novel problem, we develop three different trainable models based on RNN and Generative Adversarial Network (GAN). The models (bucket, filter bank, and end-to-end) differ in how much expert knowledge is encoded, with the most general version being purely end-to-end. To train these systems, we use Amazon Mechanical Turk to collect textual descriptions for around 2000 image pairs sampled from several datasets. Experimental results evaluated on our dataset validate our approaches. In addition, given that the filter bank model is a good compromise between generality and performance, we investigate it further by replacing RNN with Graph RNN, and show that Graph RNN improves performance. To the best of our knowledge, this is the first computational photography work on global image editing that is purely based on free-form textual instructions.

Jason D. Williams, Kavosh Asadi, Geoffrey Zweig

End-to-end learning of recurrent neural networks (RNNs) is an attractive solution for dialog systems; however, current techniques are data-intensive and require thousands of dialogs to learn simple behaviors. We introduce Hybrid Code Networks (HCNs), which combine an RNN with domain-specific knowledge encoded as software and system action templates. Compared to existing end-to-end approaches, HCNs considerably reduce the amount of training data required, while retaining the key benefit of inferring a latent representation of dialog state. In addition, HCNs can be optimized with supervised learning, reinforcement learning, or a mixture of both. HCNs attain state-of-the-art performance on the bAbI dialog dataset, and outperform two commercially deployed customer-facing dialog systems.

Kavosh Asadi, Jason D. Williams

Representing a dialog policy as a recurrent neural network (RNN) is attractive because it handles partial observability, infers a latent representation of state, and can be optimized with supervised learning (SL) or reinforcement learning (RL). For RL, a policy gradient approach is natural, but is sample inefficient. In this paper, we present 3 methods for reducing the number of dialogs required to optimize an RNN-based dialog policy with RL. The key idea is to maintain a second RNN which predicts the value of the current policy, and to apply experience replay to both networks. On two tasks, these methods reduce the number of dialogs/episodes required by about a third, vs. standard policy gradient methods.

Jason D. Williams, Geoffrey Zweig

This paper presents a model for end-to-end learning of task-oriented dialog systems. The main component of the model is a recurrent neural network (an LSTM), which maps from raw dialog history directly to a distribution over system actions. The LSTM automatically infers a representation of dialog history, which relieves the system developer of much of the manual feature engineering of dialog state. In addition, the developer can provide software that expresses business rules and provides access to programmatic APIs, enabling the LSTM to take actions in the real world on behalf of the user. The LSTM can be optimized using supervised learning (SL), where a domain expert provides example dialogs which the LSTM should imitate; or using reinforcement learning (RL), where the system improves by interacting directly with end users. Experiments show that SL and RL are complementary: SL alone can derive a reasonable initial policy from a small number of training dialogs; and starting RL optimization with a policy trained with SL substantially accelerates the learning rate of RL.