brendan o’connor

brendan o’connor

@brenocon.bsky.social

natural language processing, social science, umass, western mass http://brenocon.com he/him

2.2K followers51 following198 posts

brendan o’connor reposted

Looking forward to the Text as Data conference on Monday at UC Berkeley! I will be presenting our new paper "Multi-Perspective LLM Annotations for Valid Analyses in Subjective Tasks" which is a joint effort with Navya Mehrotra and @gligoric.bsky.social. 🧵

brendan o’connor reposted

I'm currently located in the bay area, so I'm excited to be at #tada2026 tomorrow presenting a poster on synthetic data augmentation for domain-restricted humanities language modeling and its implications for lexical leakage. I really enjoyed my last Text as Data experience, happy to see it back

brendan o’connor reposted

#TADA2026 is now up on lea.ac! Full schedule, filterable by people you follow, a custom feed, and more! (This is not part of the official TADA organization, just an experiment. Try it out and let us know what you think! Also a nice way to follow along virtually if you can't attend in person!)

Adam Visokay@avisokay.bsky.social

Looking forward to the Text as Data conference on Monday at UC Berkeley! I will be presenting our new paper "Multi-Perspective LLM Annotations for Valid Analyses in Subjective Tasks" which is a joint effort with Navya Mehrotra and @gligoric.bsky.social. 🧵

This year UMass Amherst CICS will be hiring tenure-track faculty in natural language processing! Job ad to be posted soon. For anyone at #TADA2026 or #COLM2026 this week, I or my colleague Hamed Zamani would love to chat or answer questions about it - just say hi or send me an email to meet!

brendan o’connor reposted

Killed. NSF’s SBE was killed, in violation of law. The President does not have the power to choose to kill programs at a whim by defunding them. The imperial presidency was a mistake and we are in a separation of powers constitutional crisis.

This post is unavailable.

The deadline has been extended to Monday, August 3 --

Denis@dpeskoff.bsky.social

Do you work across computational methods, social sciences, and the humanities? Submit to Text as Data 2026! 📄 One-page submissions 🔓 Non-archival ⏰ Due August 1 📍 October 5 @UCBerkeley tada2026.org

brendan o’connor reposted

Text as Data has been a wonderful, long-running, non-archival workshop for empirical research at the intersection of AI and social science (especially work involving text). After a few years off, it will be happening again this year as a one-day event in early October!

Denis@dpeskoff.bsky.social

Do you work across computational methods, social sciences, and the humanities? Submit to Text as Data 2026! 📄 One-page submissions 🔓 Non-archival ⏰ Due August 1 📍 October 5 @UCBerkeley tada2026.org

TADA is back!!! Abstract submissions very soon, take a look!

Denis@dpeskoff.bsky.social

Do you work across computational methods, social sciences, and the humanities? Submit to Text as Data 2026! 📄 One-page submissions 🔓 Non-archival ⏰ Due August 1 📍 October 5 @UCBerkeley tada2026.org

brendan o’connor reposted

Text as Data is happening at Berkeley right before COLM. Please share!

Denis@dpeskoff.bsky.social

Do you work across computational methods, social sciences, and the humanities? Submit to Text as Data 2026! 📄 One-page submissions 🔓 Non-archival ⏰ Due August 1 📍 October 5 @UCBerkeley tada2026.org

brendan o’connor reposted

Replying to @davidjurgens.bsky.social

With my ACL PC abilities, I wanted to characterize who these folks are, whether they're actually "outsiders", whether their papers are much less likely to be accepted, is this growth just LLM slop? The whole analsis is here medium.com/@jurgens_245...

medium.comIs the ACL Rolling Review actually broken?ARR is broken! ARR is being overwhelmed with papers from outsiders! LLM-generated slop is ruining our peer review! There are not enough…

brendan o’connor reposted

Scholarly folks who have ever received federal funding: today is the deadline for public comment on proposed revisions that will decimate our national research capacity and replace expertise with political whim. Please consider throwing in your 2¢. See below for some useful links and encouragement.

brendan o’connor reposted

1/ New @Nature! We study how powerful institutions shape the information environment for LLMs. Commercial LLM training is opaque, so we trace a path from state-coordinated media -> training data -> model responses.

terrible (...we go there a lot)

The New York Times@nytimes.com

Hampshire College, a small liberal arts school in Massachusetts, will permanently shutter, joining an epidemic of college closures. It's the alma mater of the filmmaker Ken Burns, who made his first documentary movie as a student there.

It's crazy and terrible to eliminate SBE, and especially the idea NSF would do it in violation of Congressional direction. Even if you only cared about AI, it's important. SBE's support of opinion data has crucially helped AI alignment research; for example, data from GSS for LLM eval & modeling.

This post is unavailable.

More broadly, there are many research questions about AI that use findings/data from the social sciences, and since their funding is so low compared to computer science, federal funding goes a *long* way there. SBE is way smaller than CISE, but accounts for a majority of socsci funding in the U.S.

brendan o’connor reposted

I don't think people fully appreciate how apocalyptic things are for US science. I haven't had any new funding since 2024, but I'm still ok since typical grants are for three years. This means next year I will be completely out of funding and will have to fire everyone in the lab. It's not great.

Jeremy Berg@jeremymberg.bsky.social

NSF Update through March 13, 2026 1/2

brendan o’connor reposted

I am flabbergasted I am by how much vibe coding has expanded my capacities as a scientist and teacher. In the last few weeks, I've mocked up class demos of a live turing test, generated cross-references for an encyclopedia, and prototyped new tablet tasks for developmental psych. It's wild.

brendan o’connor reposted

brendan o’connor reposted

The next edition of the NLP+CSS will be at ACL 2026! It includes an open-ended shared task (work with the Opioid Industry Documents Archive) with travel grants as prizes!

nlpandcss.bsky.social@nlpandcss.bsky.social

✨The NLP+CSS workshop is returning to ACL 2026!✨ And this year, we have a new shared task with prizes! Website/CfP: sites.google.com/site/nlpandc... Deadlines: March 5 (direct), March 24 (pre-reviewed ARR) #NLProc #CompSocialSci #ComputationalSocialScience #ACL2026NLP @aclmeeting.bsky.social

i agree, huge change. also as a reviewer, it used to be other reviewers on a paper were non-anonymous. when i was junior, i found this motivating and conducive to learning & community-building - i might see a review i admired, then learn more about the reviewer's work and their way of thinking.

David Mimno@dmimno.bsky.social

One thing that has changed: 10-15 years ago as an AC I personally knew 19 out of 20 reviewers for my pool. Now it’s unusual if I recognize one name. That kind of functional, internal anonymity breeds laziness.

and of course i'd want my review to not be bad since they'd see it and (potentially) know who i was! the argument for full anonymity is this of course can lead to groupthink. true, but i feel motivation & whatever means can get more high-quality reviews is the most important desideratum now

and, whatever mechanisms that can aid *mentorship* and training of more junior reviewers. leading by example is one such avenue. though, like everything, it could be the case that sheer scale destroys everything. if experienced reviewers are a small enough %age, nothing works well

also they had REU grants specifically right around the time he worked there as an undergraduate programmer (reported as 1991-1994). i dunno if his personal salary/expenses were directly paid by an NSF grant, but the existence of the project & student staff was deeply tied to NSF funding streams

brendan o’connor reposted

Great connections between our linguistics department and the CICS NLP program (including Katrin Erk now in a joint appointment). Also happy to communicate with prospective applicants.

brendan o’connor@brenocon.bsky.social

Here at UMass Amherst CICS, we’re searching for TT faculty in NLP – see the link from www.cics.umass.edu/about/employ... I’m happy to answer questions of course, too!

the way this is going the only trustworthy information will be from in-person interviews. for annotation projects i find it much easier to trust small numbers of expert or local students annotators, compared to broad-audience online recruiting

Josh McCrain@joshmccrain.bsky.social

new paper by Sean Westwood: With current technology, it is impossible to tell whether survey respondents are real or bots. Among other things, makes it easy for bad actors to manipulate outcomes. No good news here for the future of online-based survey research