Wouldn’t it solve so many problems in the world if everyone just spoke the same language? Not so fast! 

Lingthusiasm is a brand-new podcast that’s enthusiastic about linguistics, hosted by Lauren Gawne of Superlinguo​ and Gretchen McCulloch of All Things Linguistic.

In this first episode of Lingthusiasm, ​Gretchen and Lauren discuss the “one language equals peace” fallacy, and whether speaking the same words means that people will necessarily agree with each other (spoiler: no). But the history of how people have tried is still really interesting, from constructed and symbolic communication like Blissymbols and emoji to the way astronauts communicate in the high stakes environment of the International Space Station.

For links to things mentioned during this episode visit http://lingthusiasm.com/post/154520059101/lingthusiasm-episode-1-speaking-a-single-language

Listen to bonus episodes, suggest future topics, and help keep the show ad-free by supporting us on Patreon: http://patreon.com/lingthusiasm

01: Speaking a single language won’t bring about world peace

There are some pretty funny quotes of historical people complaining about kids back then doing linguistic things that now seem totally unremarkable. So let’s cut to the chase and celebrate linguistic innovation while it’s happening. 

In episode 7 of Lingthusiasm, your hosts Lauren Gawne and Gretchen McCulloch explore how far back we can trace complaints about the language of Kids These Days, why linguistic discrimination is harmful, and why “be like”, hyperbolic “literally”, and other modern innovations are actually signs of something awesome. 

We also announce a Patreon to keep the podcast sustainable! You can support us there to listen to a bonus episode about swearing and future monthly bonus content, and help decide the topics of future episodes. Even if you’re not sure about pledging, do check out the Patreon goals to see some of our future plans and our video to see a totally realistic and not at all staged cameo from our producer. Check out http://patreon.com/lingthusiasm 

For more information visit the show page: http://lingthusiasm.com/post/159796192161/lingthusiasm-episode-7-kids-these-days-arent

07: Kids these days aren’t ruining language

https://www.semanticscholar.org/paper/Design-Challenges-for-Entity-Linking-Ling-Singh/aa2a7ac7bfa9a0201d4faddd4e7bb26299a5e0be

06 - Design Challenges for Entity Linking

https://www.semanticscholar.org/paper/Capturing-Semantic-Similarity-for-Entity-Linking-w-Francis-Landau-Durrett/1c9aca60f7ac5edcceb73d612806704a7d662643

07 - Capturing Semantic Similarity for Entity Linking with Convolutional Neural Networks

The recently-hyped paper that got "superhuman" performance on FAIR's CLEVR dataset.

https://arxiv.org/abs/1706.01427

20 - A simple neural network module for relational reasoning

How do we get from knowing words to making brand-new sentences out of them? In episode 9 of Lingthusiasm, your hosts Gretchen McCulloch and Lauren Gawne talk about how words form groups with other words: constituency. 

Once you start looking for it, constituency is everywhere: in ambiguous sentences like “time flies like an arrow; fruit flies like a banana”, in remixed films like “Of Oz The Wizard”, and even internet dog memes.

This month’s Patreon bonus was the backstory about the linguistics of the doggo meme and its connection to Australian slang, which grew out of this NPR article about doggo. You can get access to it and previous bonuses about swearing, teaching yourself linguistics, and explaining linguistics to employers by supporting Lingthusiasm on Patreon.

For more information, and links to things mentioned in this episode, visit the show page: http://lingthusiasm.com/post/161859273886/lingthusiasm-episode-9-the-bridge-between-words

09: The bridge between words and sentences - Constituency

Some linguists work with multiple languages, while others focus on just one. But for many people, learning a language after early childhood is the thing that first gets them curious about how language works in general and all the things in their native language(s) that they take for granted. 

In episode 10 of Lingthusiasm, your hosts Gretchen McCulloch and Lauren Gawne talk about how learning languages can feed into linguistics and vice versa. We also explore the power dynamics that affect learning languages, and the importance of learning about the rules of interaction as well as the rules of grammar.

This month’s Patreon bonus was about hypercorrection, where you try so hard to follow a linguistic rule that you end up overshooting. You can get access to it and previous bonuses about the doggo meme, swearing, teaching yourself linguistics, and explaining linguistics to employers by supporting Lingthusiasm on Patreon. http://patreon.com/lingthusiasm

Gretchen’s new recorder in this episode is thanks to the support of our patrons! 

For more information, and links to things mentioned in this episode, visit the show page: http://lingthusiasm.com/post/163226798571/lingthusiasm-episode-10-learning-languages

10: Learning languages linguistically

EMNLP 2017 paper by Nitish Gupta, Sameer Singh, and Dan Roth.

Nitish comes on to talk to us about his paper, which presents a new entity linking model that both unifies prior sources of information into a single neural model, and trains that model in a domain-agnostic way, so it can be transferred to new domains without much performance degradation.

https://www.semanticscholar.org/paper/Entity-Linking-via-Joint-Encoding-of-Types-Descrip-Gupta-Singh/a66b6a3ac0aa9af6c178c1d1a4a97fd14a882353

33 - Entity Linking via Joint Encoding of Types, Descriptions, and Context, with Nitish Gupta

Paper by Alex Wang, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel R. Bowman.

Sam comes on to tell us about GLUE.  We talk about the motivation behind setting up a benchmark framework for natural language understanding, how the authors defined "NLU" and chose the tasks for this benchmark, a very nice diagnostic dataset that was constructed for GLUE, and what insight they gained from the experiments they've run so far.  We also have some musings about the utility of general-purpose sentence vectors, and about leaderboards.

https://www.semanticscholar.org/paper/GLUE%3A-A-Multi-Task-Benchmark-and-Analysis-Platform-Wang-Singh/a2054eff8b4efe0f1f53d88c08446f9492ae07c1

67 - GLUE: A Multi-Task Benchmark and Analysis Platform, with Sam Bowman

NAACL 2018 paper, by Rachel Rudinger, Aaron Steven White, and Benjamin Van Durme

Rachel comes on to the podcast, telling us about what factuality is (did an event happen?), what datasets exist for doing this task (a few; they made a new, bigger one), and how to build models to predict factuality (turns out a vanilla biLSTM does quite well).  Along the way, we have interesting discussions about how you decide what an "event" is, how you label factuality (whether something happened) on inherently uncertain text (like "I probably failed the test"), and how you might use a system that predicts factuality in some end task.

https://www.semanticscholar.org/paper/Neural-models-of-factuality-Rudinger-White/4d62a1e7819f9e3f8c837832c66659db5a6d9b37

68 - Neural models of factuality, with Rachel Rudinger

In this episode, we invite Hal Daumé to continue the discussion on reinforcement learning, focusing on how it has been used in NLP. We discuss how to reduce NLP problems into the reinforcement learning framework, and circumstances where it may or may not be useful. We discuss imitation learning, roll-in and roll-out, and how to approximate an expert with a reference policy. 

DAgger: https://www.semanticscholar.org/paper/A-Reduction-of-Imitation-Learning-and-Structured-to-Ross-Gordon/17eddf33b513ae1134abadab728bdbf6abab2a05?navId=citing-papers
RESLOPE: http://legacydirs.umiacs.umd.edu/~hal/docs/daume18reslope.pdf

75 - Reinforcement / Imitation Learning in NLP, with Hal Daumé III

In this episode, we invite John Hewitt to discuss his take on how to probe word embeddings for syntactic information. The basic idea is to project word embeddings to a vector space where the L2 distance between a pair of words in a sentence approximates the number of hops between them in the dependency tree. The proposed method shows that ELMo and BERT representations, trained with no syntactic supervision, embed many of the unlabeled, undirected dependency attachments between words in the same sentence.

Paper: https://nlp.stanford.edu/pubs/hewitt2019structural.pdf
GitHub repository: https://github.com/john-hewitt/structural-probes
Blog post: https://nlp.stanford.edu/~johnhew/structural-probe.html
Twitter thread: https://twitter.com/johnhewtt/status/1114252302141886464
John's homepage: https://nlp.stanford.edu/~johnhew/

88 - A Structural Probe for Finding Syntax in Word Representations, with John Hewitt

In this episode, we invite Zhou Yu to give an overview of dialogue systems. We discuss different types of dialogue systems (task-oriented vs. non-task-oriented), the main building blocks and how they relate to other research areas in NLP, how to transfer models across domains, and the different ways used to evaluate these systems. Zhou also shares her thoughts on exciting future directions such as developing dialogue methods for non-cooperative environments (e.g., to negotiate prices) and multimodal dialogue systems (e.g., using video as well as audio/text).

Zhou Yu's homepage: http://zhouyu.cs.ucdavis.edu/

89 - Dialog Systems, with Zhou Yu

In this episode, we invite Jonathan Berant to talk about executable semantic parsing. We discuss what executable semantic parsing is and how it differs from related tasks such as semantic dependency parsing and abstract meaning representation (AMR) parsing. We talk about the main components of a semantic parser, how the formal language affects design choices in the parser, and end with a discussion of some exciting open problems in this space.

Jonathan Berant's homepage: http://www.cs.tau.ac.il/~joberant/

91 - (Executable) Semantic Parsing, with Jonathan Berant

In this episode, we invite David Bamman to give an overview of computational humanities. We discuss examples of questions studied in computational humanities (e.g., characterizing fictionality, assessing novelty, measuring the attention given to male vs. female characters in the literature). We talk about the role NLP plays in addressing these questions and how the accuracy and biases of NLP models can influence the results. We also discuss understudied NLP tasks which can help us answer more questions in this domain such as literary scene coreference resolution and constructing a map of literature geography.

David Bamman's homepage: http://people.ischool.berkeley.edu/~dbamman/
LitBank dataset: https://github.com/dbamman/litbank

92 - Computational Humanities, with David Bamman

**The podcast is currently on hiatus. For more active NLP content, check out the Holistic Intelligence Podcast linked below.**

Welcome to the NLP highlights podcast, where we invite researchers to talk about their work in various areas in natural language processing. All views expressed belong to the hosts/guests, and do not represent their employers.

67 - GLUE: A Multi-Task Benchmark and Analysis Platform, with Sam Bowman

NLP Highlights