Vision-and-Language Navigation wayfinding agents can be enhanced by exploiting automatically generated navigation instructions. However, existing instruction generators have not been comprehensively evaluated, and the automatic evaluation metrics used to develop them have not been validated. Using human wayfinders, we show that these generators perform on par with or only slightly better than a template-based generator and far worse than human instructors. Furthermore, we discover that BLEU, ROUGE, METEOR and CIDEr are ineffective for evaluating grounded navigation instructions. To improve instruction evaluation, we propose an instruction-trajectory compatibility model that operates without reference instructions. Our model shows the highest correlation with human wayfinding outcomes when scoring individual instructions. For ranking instruction generation systems, if reference instructions are available we recommend using SPICE.

PanGEA, the Panoramic Graph Environment Annotation toolkit, is a lightweight toolkit for collecting speech and text annotations in photo-realistic 3D environments. PanGEA immerses annotators in a web-based simulation and allows them to move around easily as they speak and/or listen. It includes database and cloud storage integration, plus utilities for automatically aligning recorded speech with manual transcriptions and the virtual pose of the annotators. Out of the box, PanGEA supports two tasks – collecting navigation instructions and navigation instruction following – and it could be easily adapted for annotating walking tours, finding and labeling landmarks or objects, and similar tasks. We share best practices learned from using PanGEA in a 20,000 hour annotation effort to collect the Room-Across-Room dataset. We hope that our open-source annotation toolkit and insights will both expedite future data collection efforts and spur innovation on the kinds of grounded language tasks such environments can support.

We present WHERE ARE YOU? (WAY), a dataset of ~6k dialogs in which two humans – an Observer and a Locator – complete a cooperative localization task. The Observer is spawned at random in a 3D environment and can navigate from first-person views while answering questions from the Locator. The Locator must localize the Observer in a detailed top-down map by asking questions and giving instructions. Based on this dataset, we define three challenging tasks: Localization from Embodied Dialog or LED (localizing the Observer from dialog history), Embodied Visual Dialog (modeling the Observer), and Cooperative Localization (modeling both agents). In this paper, we focus on the LED task – providing a strong baseline model with detailed ablations characterizing both dataset biases and the importance of various modeling choices. Our best model achieves 32.7% success at identifying the Observer’s location within 3m in unseen buildings, vs. 70.4% for human Locators.

We introduce Room-Across-Room (RxR), a new Vision-and-Language Navigation (VLN) dataset. RxR is multilingual (English, Hindi, and Telugu) and larger (more paths and instructions) than other VLN datasets. It emphasizes the role of language in VLN by addressing known biases in paths and eliciting more references to visible entities. Furthermore, each word in an instruction is time-aligned to the virtual poses of instruction creators and validators. We establish baseline scores for monolingual and multilingual settings and multitask learning when including Room-to-Room annotations (Anderson et al., 2018). We also provide results for a model that learns from synchronized pose traces by focusing only on portions of the panorama attended to in human demonstrations. The size, scope and detail of RxR dramatically expands the frontier for research on embodied language agents in photorealistic simulated environments.

In recent years, the natural language processing community has moved away from task-specific feature engineering, i.e., researchers discovering ad-hoc feature representations for various tasks, in favor of general-purpose methods that learn the input representation by themselves. However, state-of-the-art approaches to disfluency detection in spontaneous speech transcripts currently still depend on an array of hand-crafted features, and other representations derived from the output of pre-existing systems such as language models or dependency parsers. As an alternative, this paper proposes a simple yet effective model for automatic disfluency detection, called an auto-correlational neural network (ACNN). The model uses a convolutional neural network (CNN) and augments it with a new auto-correlation operator at the lowest layer that can capture the kinds of “rough copy” dependencies that are characteristic of repair disfluencies in speech. In experiments, the ACNN model outperforms the baseline CNN on a disfluency detection task with a 5% increase in f-score, which is close to the previous best result on this task.

Because obtaining training data is often the most difficult part of an NLP or ML project, we develop methods for predicting how much data is required to achieve a desired test accuracy by extrapolating results from models trained on a small pilot training dataset. We model how accuracy varies as a function of training size on subsets of the pilot data, and use that model to predict how much training data would be required to achieve the desired accuracy. We introduce a new performance extrapolation task to evaluate how well different extrapolations predict accuracy on larger training sets. We show that details of hyperparameter optimisation and the extrapolation models can have dramatic effects in a document classification task. We believe this is an important first step in developing methods for estimating the resources required to meet specific engineering performance targets.

A long-term goal of AI research is to build intelligent agents that can see the rich visual environment around us, communicate this understanding in natural language to humans and other agents, and act in a physical or embodied environment. To this end, recent advances at the intersection of language and vision have made incredible progress – from being able to generate natural language descriptions of images/videos, to answering questions about them, to even holding free-form conversations about visual content! However, while these agents can passively describe images or answer (a sequence of) questions about them, they cannot act in the world (what if I cannot answer a question from my current view, or I am asked to move or manipulate something?). Thus, the challenge now is to extend this progress in language and vision to embodied agents that take actions and actively interact with their visual environments. To reduce the entry barrier for new researchers, this tutorial will provide an overview of the growing number of multimodal tasks and datasets that combine textual and visual understanding. We will comprehensively review existing state-of-the-art approaches to selected tasks such as image captioning, visual question answering (VQA) and visual dialog, presenting the key architectural building blocks (such as co-attention) and novel algorithms (such as cooperative/adversarial games) used to train models for these tasks. We will then discuss some of the current and upcoming challenges of combining language, vision and actions, and introduce some recently-released interactive 3D simulation environments designed for this purpose.

Existing image captioning models do not generalize well to out-of-domain images containing novel scenes or objects. This limitation severely hinders the use of these models in real world applications dealing with images in the wild. We address this problem using a flexible approach that enables existing deep captioning architectures to take advantage of image taggers at test time, without re-training. Our method uses constrained beam search to force the inclusion of selected tag words in the output, and fixed, pretrained word embeddings to facilitate vocabulary expansion to previously unseen tag words. Using this approach we achieve state of the art results for out-of-domain captioning on MSCOCO (and improved results for in-domain captioning). Perhaps surprisingly, our results significantly outperform approaches that incorporate the same tag predictions into the learning algorithm. We also show that we can significantly improve the quality of generated ImageNet captions by leveraging ground-truth labels.