Kuleen Sasse

PhD @JHUCompSci | NSF GRF | Prev. @geodatascience @JHUAPL

[Kuleen Sasse @oldbayes]

📣 New paper accepted to #ACL2025 Main!📣

We are able to detect political dog whistles in controlled environments but can NLP models discover current and novel dog whistles in the wild?

Across multiple NLP techniques, our new paper finds a clear answer: no!
🧵

[Dominik Stammbach @dominsta_nlp]

How can AI support public defense? We have worked on this question for the last two years, and have (1) built an AI retrieval tool for the NJ Office of the Public Defender and (2) conducted qualitative interviews with defenders. Our main insights 👇(1/12)

[Washington Examiner @dcexaminer]

OP-ED: The uncomfortable truth behind AI debate — and a bipartisan solution https://t.co/GTRCOubV6l

[Balogun @balogun]

My debut World Cup… it hurts to wait 4 years to compete at the highest level our sport has to offer. I want to say sorry to our fans it was not good enough when it mattered most and we let you down

Soccer in America will only become bigger the belief, the talent, and the passion is continually growing and I know the best days are in front of us, the future belongs to those who never stop believing, this moment will fuel us. We will be back

Why not us?

For the nation. For the flag.
🇺🇸🦅
B

[William Rudman @WilliamRudmanjr]

@MichalGolov and I will be presenting our work locating causal mechanisms of “prompt induced hallucinations” in VLMs at #ACL2026 tomorrow at 11am in the Grand Hall!
📖: https://t.co/uql7dGybiH

[Kristina Gligorić @krisgligoric]

P-hacking is a long-standing problem in science, and LLMs make it worse: as tools for annotation or LLM-as-a-judge, they allow tuning parameters until a desired result appears.

We propose a way to mitigate this problem, and conduct a preregistered study of its effectiveness!

[Tianjian Li @tli104]

🚨 New paper: "Self-Compacting Language Model Agents"

LM agents build up long traces of reasoning and tool calls. As the trace grows, old mistakes and stale info stick around and anchor everything that follows. We ask: can the model itself decide when to clean up?

[Nick Caputo @nickacaputo]

Excited to announce that I'll be joining the new Johns Hopkins School of Government and Policy @JHUBloombergCtr as an Assistant Professor next week!

Frontier AI governance is only getting harder and more important and I'm so excited to get to work with @ghadfield, @sethlazar, and many other brilliant minds to help build a brighter future.

Many fun projects to come!

[Nari Johnson @narijohnson]

🔬Science can play a critical role in guiding policy interventions. How can we make our research legible to the policymakers who work on AI?

Our tutorial "From FAccT to Policy Impact" explores how we can translate our research in a specific format: the policy memo! #FAccT2026

[Gillian Hadfield @ghadfield]

.@sethlazar, @nickacaputo, and I are launching a new AGI Governance Fellowship at Johns Hopkins: three weeks in DC this September, in person, for early-career people already in AI governance. A capstone, not an on-ramp. Applications open, mid-July for full consideration.

Apply here: https://t.co/TNU4Dpu4sU

[Seth Lazar @sethlazar]

Announcing a 3 wk AGI Governance Fellowship at @JHUBloombergCtr School of Government and Policy.

Join me, @ghadfield, @nickacaputo and guests for an intensive schedule of deep dives into AI governance at the frontier, covering topics such as societal resilience, AGI and democracy, and the new institutions that governing in/with AGI will demand; we will range from foundational philosophical questions to the tip of the legislative spear.

The arrival of AGI would be challenging enough if our ~250yo institutions were in full health. That they aren’t both increases the scale of the challenge, and creates a unique opportunity to imagine the institutions that will carry us through the next 1/4 millennium.

We are looking for ~20 fellows, aiming to attract and shape the people who will start to build those institutions: future leaders in AGI governance. We expect you to come from many different backgrounds, so hope this call will be shared far and wide. Details, eligibility, and how to apply here: https://t.co/k4IdSSpmrM

Governing the AI transition is going to be a central focus of our new school, and we have many more initiatives coming. And we will be hiring in many different kinds of role, from tenured and tenure track faculty to research scientists to ops and comms and students and beyond.

[Kenneth Marino @Kenneth_Marino]

Absolutely insane graph here. I’m sorry, if you wrote 40 papers for ACL, no you didn’t. @CVPR put a cap and a bunch of other reforms to try to head off the paper-pocalypse. ACL was caught flat-footed and now it sounds like they’re getting absolutely dogpiled by angry reviewers.

[Ziyang Huang @zyh_ai]

Can coding agents actually reproduce scientific findings — not just write code, but rerun the science?

We built AutoMat to find out. 🧪🤖

AutoMat is a benchmark of 85 expert-curated reproduction tasks from computational materials science.

[William Rudman @WilliamRudmanjr]

Hypothesis -> experiments -> analysis -> conclusions. LLMs are great at writing code and conducting experiments.

But there’s a weakness in their ability to propose statistical models and evaluate their fit.

Enter VESTA: Visual Exploration with Statistical Tool Agents.

[Jonathan Ivey @Jonathan_Ivey8]

I just finished the first year of my PhD. Rather than just share my own perspective, I sat down with 7 other students in my cohort and talked with them about their experiences in depth. In my new post, I share some of our biggest lessons: https://t.co/QVKu105Lq4

[Kuleen Sasse @oldbayes]

This might be the highest alpha thing you read all week!

[Jonathan Ivey @Jonathan_Ivey8]

I just finished the first year of my PhD. Rather than just share my own perspective, I sat down with 7 other students in my cohort and talked with them about their experiences in depth. In my new post, I share some of our biggest lessons: https://t.co/QVKu105Lq4

[Daniel Khashabi 🕊️ @DanielKhashabi]

Happy to share that Ben Van Durme @ben_vandurme has been promoted to Full Professor, effective immediately!

Congratulations on this well-deserved recognition!

[Kuleen Sasse @oldbayes]

Winner

[Jonathan Ivey @Jonathan_Ivey8]

What makes a good participant response in qualitative research? In our new paper, we investigate this by analyzing 16,940 responses across 343 interviews to see which characteristics of a response make it most likely to contribute to a study’s findings.

Here’s what we found …

[Jonathan Ivey @Jonathan_Ivey8]

What makes a good participant response in qualitative research? In our new paper, we investigate this by analyzing 16,940 responses across 343 interviews to see which characteristics of a response make it most likely to contribute to a study’s findings.

Here’s what we found …

[Mark Dredze @mdredze]

Huge congratulations to my PhD student @isabelcachola on defending her thesis, "Methods for the generation and evaluation of user-centric scientific summarization"! 🎓🎉🍾

Impressively, she created a jacket for the event made entirely from fabric posters of her work! 🧵🧥📄✨

[Jack Jingyu Zhang @jackjingyuzhang]

Real-world agents juggle instructions from skill files, tools, other agents, ... each with different trust levels. When these conflict, can models reliably prioritize the most trusted one?

Our ManyIH-Bench🪜 finds that even frontier models like GPT-5.4 only get ~40% accuracy! 👇