I’m thrilled that this Human-Centered MT paper was recognized with an outstanding paper award at #EMNLP2023. Congratulations to lead authors Nikita Mehandru (@ucberkeley iSchool) and Sweta Agrawal (UMD CLIP Instituto Superior Técnico) for making this interdisciplinary collaboration a success!

Awesome work! Congrats Nikita Mehandru Sweta Agrawal et al!!!!

Congrats to Masha et al on this best industry paper award! (Masha’s a Berkeley School of Information MIMS alum!)

🚨NLP+CSS workshop is back and will be at NAACL 2024! Paper submission deadline: March 24 sites.google.com/site/nlpandcss/ Organizing team: Anjalie Field Dallas Card @dirk_hovy and myself

New preprint! 📜 We investigate how ten “quality” and English langID filters, drawn from prior lit on LLM pretraining data curation pipelines, affect webpages linked to self-descriptions of their creators. Paper: arxiv.org/abs/2401.06408 Data: huggingface.co/datasets/allen… 🧵(1/6)

Very excited to announce the launch of our citizen science initiative "The Lives of Literary Characters" hosted The Zooniverse. This is the first ever literary citizen science project that aims to promote story understanding. A Thread 🧵 zooniverse.org/projects/citiz…

Join us on Monday, 2/26 at 4:30 pm for a lecture by David Bamman: The Promise and Peril of Large Language Models for Cultural Analytics. RSVP: forms.gle/by1m6xHzTLhjJQ… More info: cdh.princeton.edu/events/2024/02… Co-sponsored by Princeton PLI.

I’m headed to NAACL to present this paper! I’m around mostly Sunday evening thru Tuesday. This fall I’ll be doing some thinking about what to do after my PhD; if you have advice/thoughts about this definitely chat with me!

Hey NLPals, I'll be at #NAACL2024 this upcoming week! Let's chat about sociocultural NLP, what it means to study culture, and finding variation in unusual places (like memes!) I'll be presenting this memes paper at the first poster session.

Looking forward to seeing people at #NAACL2024 this week! Today, be sure to check out Naitian Zhou's poster on the sociolinguistics of memes (11am) and Lucy Li's talk on concepts of fairness in NLG systems at 2:36pm (ethics/bias/fairness 1)

For anyone at #NAACL2024 considering lucha libre, I can attest it was spectacular (though that may be influenced by attending with my 9yo)

See Naitian at poster #5!

It’s an extraordinary pleasure and honor to teach alongside David Bamman and his wonderful students of NLP, now doubly so to have my small part recognized by Berkeley School of Information & UC Berkeley.

Well deserved, congrats Kent!

Big congrats to Kent K. Chang for passing his qualifying exam today! Lots of super exciting work on measuring social interactions in culture in the pipeline --

My group just finished up a new paper that I'm excited to get out into the world: "On Classification with Large Language Models in Cultural Analytics" (to be published at CHR): github.com/bamman-group/c…. More info here! bsky.app/profile/dbamma…

How might one do classification in the era of LLMs for humanities research? 🤔 David Bamman, Kent K. Chang, Naitian Zhou & I apply LLMs on ten tasks from prior cultural analytics lit. Larger LMs are competitive w/ older methods on established tasks, but perform less well on new ones.

In cultural analytics, accuracy is often not the only (or even primary) objective. Here, we explore the myriad ways CA uses classification, how LLMs compare to other commonly used methods, and how they might enable new approaches to sensemaking from text data.

Lucy is a rock star and you should all hire her!

From David Bamman (co-authored w/Rachael G. S.) new paper today relying on #copyright exemption for decrypting DVDs to conduct #textdatamining. Using the exemption, authors built a collection of film to measure representation for gender and race/ethnicity pnas.org/doi/10.1073/pn…