Catch up with your colleagues at the Welcome Reception! It will be held immediately following the tutorials at the Westin Bayshore Hotel, Sunday, July 30th, in the Bayshore Grand Ballroom (the conference venue). Refreshments and a light dinner will be provided and a cash bar will be available.

Computational linguistics is a booming field and our association is flourishing with it. As our conferences grow larger and the pace of publishing quickens, there is a constant need to reflect on strategies that will allow us to prosper and grow even stronger in the future. In my presidential address, I will focus on three topics that I think require our attention. The first is equity and diversity, where the ACL executive committee has recently launched a number of actions intended to improve the inclusiveness and diversity of our community, but where there is clearly a need to do more. The second topic is publishing and reviewing, where the landscape is changing very quickly and our current system is starting to strain under the sheer volume of submissions. In particular, there has been an active discussion recently about the pros and cons of preprint publishing and the way it interacts with our standard model for double-blind reviewing. On this topic, I will present the results of a large-scale survey organized by the ACL executive committee to learn more about current practices and views in our community, a survey that will be followed up by a panel and discussion at the ACL business meeting later in the week. The third and final topic is good science and what we can do to promote scientific methodology and research ethics, which is becoming increasingly important in a world where the role of science in society cannot be taken for granted.

The computational linguistics and natural language processing community is experiencing an episode of deep fascination with representation learning. Like many other presenters at this conference, I will describe new ways to use representation learning in models of natural language. Noting that a data-driven model always assumes a theory (not necessarily a good one), I will argue for the benefits of language-appropriate inductive bias for representation-learning-infused models of language. Such bias often comes in the form of assumptions baked into a model, constraints on an inference algorithm, or linguistic analysis applied to data. Indeed, many decades of research in linguistics (including computational linguistics) put our community in a strong position to identify promising inductive biases. The new models, in turn, may allow us to explore previously unavailable forms of bias, and to produce findings of interest to linguistics. I will focus on new models of documents and of sentential semantic structures, and I will emphasize abstract, reusable components and their assumptions rather than applications.

With over 50,000 amazing aquatic creatures, the Vancouver Aquarium Marine Science Centre provides a beautiful backdrop for our event with unforgettable galleries filled with aquatic life, plus expansive outdoor spaces.

The Vancouver Aquarium is a self-supporting, non-profit society dedicated to effecting the conservation of aquatic life through display and interpretation, education, research and direct action. Our event will help to achieve a positive effect on the marine world and directly contribute to their conservation, research and education programs.

The social event is included with your registration. We will also offer additional admissions for people accompanying participants (see registration form for pricing details).

Learn more about the social event here.

Recent years have witnessed the development of a wide range of computational tools that process and generate natural language text. Many of these have become familiar to mainstream computer users in the from of web search, question answering, sentiment analysis, and notably machine translation. The accessibility of the web could be further enhanced with applications that not only translate between different languages (e.g., from English to French) but also within the same language, between different modalities, or different data formats. The web is rife with non-linguistic data (e.g., video, images, source code) that cannot be indexed or searched since most retrieval tools operate over textual data.

In this talk I will argue that in order to render electronic data more accessible to individuals and computers alike, new types of translation models need to be developed. I will focus on three examples, text simplification, source code generation, and movie summarization. I will illustrate how recent advances in deep learning can be extended in order to induce general representations for different modalities and learn how to translate between these and natural language.