This is the main course website. There is also a course Canvas site for uploading assignments.

Class Business

Readings for Next Classes

For Tuesday (Feb. 25th)
For next Thursday (Feb 27th)
- Note: class on Thursday will be over Zoom due to a scheduled internet outage in South Hall)
Due Thursday: Large Language Models & Text-to-Image Large Models Exercise

Word Embedding

Luis Serrano, “What Are Word and Sentence Embeddings?” (2023)
Saptarashmi Bandyopadhyay et al., “Word Embedding Demo: Tutorial” (2022) — Note: The actual interactive demo accompanying this tutorial about word embeddings (or word vectors) is assigned for the second part of Practicum 5.
Nika Mavrody, Laura B. McGrath, Nichole Nomura, and Alexander Sherman, “Voice” (2021) — Read the “Abstract” and pp. 155-164.
Ryan Heuser, “Word Vectors in the Eighteenth Century” (conference proceedings abstract) (2017)

“Infinite Craft” by Neal Agarwal

Tensorflow Embedding Projector (Google AI)

Saptarashmi Bandyopadhyay et al., “Word Embedding Demo: Tutorial” (2022) — Interactive demo

Semantris game from Google AI

Practicum 5: Word Embedding Exercise

Student outputs

Randomizer

“Interpretability” and “Explainability”
Alan D. Thompson, “What’s in My AI?” (2022)
OED
- Historical Thesaurus of the Oxford English Dictionary (2009)

Using Word Embedding

Applied Use Examples from Assigned Readings

Nika Mavrody, Laura B. McGrath, Nichole Nomura, and Alexander Sherman, “Voice” (2021) — “Abstract” and pp. 155-164.

Ultimately, we show that voice, style, and genre operate in a unified vernacular critical system, that voice (along with genre) is a subcategory of style, and that voice consists of the parts of style not otherwise captured by genre.

“Voice”
“Style”
“Style – Voice = Genre”
“Style – Genre = Voice”

Ryan Heuser, “Word Vectors in the Eighteenth Century” (conference proceedings abstract) (2017)

Ryan Heuser, "Word Vectors in the Eighteeth Century" (2017), Table 1 — Table 1

Ryan Heuser, "Word Vectors in the Eighteeth Century" (2017), Table 2 — Table 2

Word Embedding Demo (by S. Bandyopadhyay et al.)

Try finding analogies in the demo.

James Lee, James, Blaine Greteman, Jason Lee, and David Eichmann, “Linked Reading: Digital Historicism and Early Modern Discourses of Race around Shakespeare’s Othello” (2018)
“Word2Vec Model of “Complexion” (EEBO Corpus from 1500 to 1623)” — Fig. 14 in James Lee, Blaine Greteman, Jason Lee, and David Eichmann. “Linked Reading: Digital Historicism and Early Modern Discourses of Race around Shakespeare’s Othello.” Journal of Cultural Analytics, 2018.

Case Study

Amendments to the U.S. Bill of Rights (National Archives transcription)

Amendment II

A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed.

Your Uses for Word Embedding?

?

Problems with Word Embedding?

Bias
- Parul Pandey, “Visualizing Bias in Data Using Embedding Projector” (2019)
“Catachresis”
- Fabian Offert, “Intuition and Epistemology of High-Dimensional Vector Space” (2021)
- Jacques Derrida, Margins of Philosophy (1982)
  - Catachresis “concerns first the violent, forced, abusive inscription of a sign, the imposition of a sign upon a meaning which did not yet have its own proper sign in language” (255)
- J. Hillis Miller, “The Figure in the Carpet” (1980):
  - “Catachresis is the name for that procedure whereby [Henry] James uses all the realistic detail of his procedure as a novelist to name in figure, abusive transfer, something else for which there is no literal name and therefore, within the convention of referentiality which the story as a realistic novel accepts, no existence” (111)