LIN 389C Schedule

Course outline Fall 2020

This semester we have one meeting per week. Each week, unless noted otherwise, we use one half of the class to discuss the topic of the week, and we use the other half for a round-table discussing people's research.


Week 1: Aug 28

  • Plan for the semester: Which topics to focus on, other activities
  • Extended round table:
    • Research results from the summer
    • Research plans for the fall, publication plans

Week 2: Sep 4: Rational Speech Acts

We are reading an introduction to Rational Speech Acts:
Goodman and Frank, Pragmatic Language Interpretation as Probabilistic Inference, Trends in CogSc 2016, Vol 20 No. 11, https://cocolab.stanford.edu/papers/GoodmanFrank2016-TICS.pdf

(If you are interested in one of the probabilistic programming languages that Goodman and colleagues have developed to use with RSA, there is an intro to RSA with WebPPL code at http://www.problang.org/chapters/01-introduction.html. WebPPL is built on top of javascript.)

Venkat will lead the discussion.

Week 3: Sep 11: Rational Speech Acts and metaphor

We are reading a paper that applies RSA to metaphor interpretation:
Kao, Bergen and Goodman, Formalizing the Pragmatics of Metaphor Understanding. CogSci conference, 2014. https://escholarship.org/content/qt09h3p4cz/qt09h3p4cz.pdf

and another paper using a similar technique:

Degen, Tessler, Goodman, Wonky worlds: Listeners revise world knowledge when utterances are odd, CogScin conference, 2015. https://cocolab.stanford.edu/papers/DegenEtAl2015-Cogsci.pdf

Gabriella will lead the discussion.

Week 4: Sep 18: Metaphor

We will hear from Gabriella about her thesis work on metaphor.

As a supplement, we're reading the following paper that introduces a new metaphor dataset:

Omnia Zayed, John P. McCrae, Paul Buitelaar, 2020. Figure Me Out: A Gold Standard Dataset for Metaphor Interpretation, LREC 2020. https://www.aclweb.org/anthology/2020.lrec-1.712.pdf


Joel pointed out this paper on the use of Bayesian approaches in psychology: Matt Jones and Brad Love, Bayesian Fundamentalismor Enlightenment? On the explanatorystatus and theoretical contributions ofBayesian models of cognition. http://matt.colorado.edu/papers/jones-love-BBS.pdf We'll briefly talk about this in class.


Week 5: Sep 25: Background knowledge in neural models

We are reading the KnowBERT paper:
Matthew E. Peters, Mark Neumann, Robert L. Logan IV, Roy Schwartz, Vidur Joshi, Sameer Singh, Noah A. Smith, Knowledge Enhanced Contextual Word Representations, EMNLP 2019, https://www.aclweb.org/anthology/D19-1005/

Ching-Da will lead the discussion.

Week 6: Oct 2: Commonsense knowledge in neural models

We are reading the paper on self-talk and language models, a very useful technique:
Vered Shwartz, Peter West,  Ronan Le Bras, Chandra Bhagavatula, and Yejin Choi, Unsupervised Commonsense Question Answering with Self-Talk, https://arxiv.org/pdf/2004.05483.pdf

Yejin will lead the discussion.

Week 7: Oct 9: NLP and ethics: data collection

Possible readings:
Joel will lead the discussion.

Week 8: Oct 16: NLP and ethics: algorithms

We could read https://www.aclweb.org/anthology/2020.acl-main.431.pdf, "Interpreting Pretrained Contextualized Representations via Reductions to Static Embeddings", which has an application to analyzing bias in contextualized embeddings

Week 9: Oct 23: Language and bias

We could  read

https://homes.cs.washington.edu/~msap/pdfs/sap2020socialbiasframes.pdf, "Social Bias Frames: Reasoning about Social and Power Implications of Language"

Week 10: Oct 30: Tentatively, knowledge graphs

Week 11: Nov 6: Tentatively, knowledge graphs

Week 12: Nov 13

Week 13: Nov 20

Week 14: Nov 26: Thanksgiving break

Week 15: Dec 4


Final course project papers due: Dec 4, 2020.