10  Week 10: Guest lecture

This week, we will have a guest lecture on large language models and algorithmic justice.

Large language models are powering a lot of the recommendation and information retrieval systems you use everyday. But given that these models are built on human-produced text, they tend to recreate the bias and stereotyping that characterize our language.

The readings for this week provide an introduction to some of these issues, as well as how we might go about addressing them.

10.1 Essential reading:

Bender et al. (2021)

Pfotenhauer et al. (2021)

10.2 Additional reading:

Solaiman and Dennison (2021)

10.3 Slides

Slides for this week are available here

Bender, Emily M., Timnit Gebru, Angelina McMillan-Major, and Shmargaret Shmitchell. 2021. “On the Dangers of Stochastic Parrots.” Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, March. https://doi.org/10.1145/3442188.3445922.

Pfotenhauer, Sebastian, Brice Laurent, Kyriaki Papageorgiou, Stilgoe, and Jack. 2021. “The Politics of Scaling.” Social Studies of Science 52 (1): 3–34. https://doi.org/10.1177/03063127211048945.

Solaiman, Irene, and Christy Dennison. 2021. “Process for Adapting Language Models to Society (PALMS) with Values-Targeted Datasets.” https://doi.org/10.48550/ARXIV.2106.10328.