Tzu-Sheng Kuo 郭子生

How can we help communities collaboratively shape policies that impact them? In our #CHI2025 paper, we present PolicyCraft, a system that supports ✨collaborative policy design✨ through case-grounded deliberation. (🧵/11)

☀️ For the 2nd year in a row, I had the honor of serving as Student Volunteer Co-Chair, leading an incredible team of over 200 SVs at @chi.acm.org My heartfelt thanks go out to our SVs, the true MVPs who work tirelessly behind the scenes. We hope all attendees had a wonderful conference!

Finally, we also had a paper with collaborator @tskuo.bsky.social at CMU on enabling collective design of community bots via a system that surfaces provocative hypothetical cases as a way to stress-test and iterate on bot policy: arxiv.org/abs/2509.25492

CSCW folks, I wanted to highlight how excited and proud I am to see work from our community (dl.acm.org/doi/10.1145/..., CSCW '24 best paper winner led by @jiachenyan.bsky.social and @mlam.bsky.social) grow and expand ambition into this Science paper. CSCW has a ton to offer the world.

Really excited about this 🔔new paper🔔 where we had a chance to leverage Change.org 's staggered rollout of a "write with AI" tool to causally (aka "once and for all") measure the impact of such tools on global platform outcomes. Summary: with AI, petition length ⬆️, homogeneity ⬆️, Outcomes ⬇️. More:

X turned on account location, and immediately turned it off again--because it showed that your "farmer in Michigan" is really a paid troll in Bangladesh. We need a social network that shows where content is from. asbruckman.medium.com/x-added-acco... With cites to @katestarbird.bsky.social

The UW Center for an Informed Public is looking for postdocs (for 2026-2028) from across diverse disciplines whose research sheds light on the challenges of our modern information environment, promotes civic health, and/or helps people/communities navigate online spaces: apply.interfolio.com/177901

Just arrived in Bergen for CSCW, where I'll present Venire! Venire is a Reddit moderation tool that uses an ML model trained on mod decision histories to identify controversial cases. It preempts inconsistent decision-making by flagging these cases for multi-mod review dl.acm.org/doi/10.1145/...

I'm headed to #UIST2025 🇰🇷 to present our 🏆 Best Paper (!!!), "Graffiti: Enabling an Ecosystem of Personalized and Interoperable Social Applications" @karger.bsky.social and David Clark doi.org/10.1145/3746...

LLM safety work often reasons over high-level policies (be helpful & polite), but must tackle on-the-ground cases (unsolicited money advice when stocks are mentioned). This can feel like driving on an unfamiliar road guided by a generic driver’s manual instead of a map. We introduce: Policy Maps 🗺️

Social media feeds today are optimized for engagement, often leading to misalignment between users' intentions and technology use. In a new paper, we introduce Bonsai, a tool to create feeds based on stated preferences, rather than predicted engagement. arxiv.org/abs/2509.10776

Crunching away on a paper for #CHI2026 about online communities, moderation, or sociolinguistics? You may be want to check out recent work led by @galenweld.bsky.social on predicting SOVC within subreddits based on the style and structure of conversations. arxiv.org/abs/2508.08596

The @acm-cscw.bsky.social Starter Pack is growing ahead of #CSCW2025 ! (go.bsky.app/SPumuMT) 🔥 A warm welcome to all of our new members 🔥: @imanm02.bsky.social @jesanovi.bsky.social @shadinz.bsky.social @jaytee-ess.bsky.social @jaewonk.bsky.social @sypark9.bsky.social New member thread 1/x

We are organizing a workshop on Algorithmic Collective Action at NeurIPS this year. As AI continues to concentrate power, we will meet in San Diego (Dec 6 or 7) for critical conversations on user coordination, labor, data protection, and community advocacy. Submissions due August 22. #NeurIPS2025

Last day of #ACL2025NLP but there's still lots to do: attend the #WikiNLP workshop, where we explore how NLP and wikipedia can help each other! We have amazing keynotes, discussions with Wikipedia editors, a panel + posters! Details: meta.wikimedia.org/wiki/NLP_for... Join us in room 2.31!

My team is organizing a workshop on information disorder at the CSCW conference (Oct 18, Bergen, Norway). We aim to convene scholars committed to understanding mis & disinfo, scams, online hate, etc. Applicants should submit 2-6 page position papers by Aug 1: cscw2025infodisorder.netlify.app

Prompting is our most successful tool for exploring LLMs, but the term evokes eye-rolls and grimaces from scientists. Why? Because prompting as scientific inquiry has become conflated with prompt engineering. This is holding us back. 🧵and new paper with @ari-holtzman.bsky.social .

Thrilled to share that I’ve successfully defended my PhD dissertation and I will be joining MIT as an Assistant Professor starting Fall 2026, with a shared appointment between Sloan and EECS! I will be recruiting 1-2 PhD students this upcoming cycle. Consider applying to MIT EECS!

Another week, another research ethics controversy. TL;DR Researchers released a public dataset of 2B+ messages from 4M+ users on 3k+ "public" Discord servers. Usernames/IDs are anonymized. But let's unpack this one... 🧵 www.404media.co/researchers-...

🚀 Curious about research on @wikipedia.org and @wikimediafoundation.org projects? Join us at #WikiWorkshop2025, May 21–22 (virtual)! 🔹 46 research talks 🔹 Live Q&A + networking Free registration: pretix.eu/wikimedia/wi...

Somewhat oddly, the Trump regime's initial moves to cut science funding went very broad. This is catalyzing solidarity & advocacy around the value of science (see comms featuring cancer cures & tech innovation). But their next move may be to try to drive a wedge b/w "good science" & "bad science".

Groups use processes to make decisions. But normal people with no training will usually run a bad process. When you just perform how you think meetings are supposed to work, without training, your meetings will be boring or bad. AND, big secret: when everyone's trained, meetings feel good.

Serving as the Student Volunteer Co-Chair at #CHI2025 @chi.acm.org has been an incredible experience! A huge thank you to the 200 student volunteers who worked tirelessly behind the scenes to help make the conference a success. Looking forward to seeing you all next year in Barcelona!

☀️ Our amazing team of student volunteers is all set and excited for the conference kickoff! #CHI2025

@kjfeng.me and @rockpang.bsky.social are kicking off the panel on sociotechnical AI governance! #CHI2025

📢📢📢 Second for Contributions! Join us at the 2nd WikiNLP Workshop on NLP for Wikipedia, co-located with #ACL2025 in Vienna. We welcome both in-person and virtual attendees, and have both archival and non-archival tracks! 🗓️ deadline: April 30 Details: meta.wikimedia.org/wiki/NLP_for...

I would love to never see the phrase "little is known about" in a paper ever again. (1) Most of the time the authors are far more confident about that statement than they should be. (2) There are lots of things we know little about because no one cares. Just not knowing isn't a good motivation.

Do you make custom feeds on Bluesky? We'd love to hear about your experiences as part of a research project at @uofwa.bsky.social @socialfutureslab.bsky.social ! Come chat with us for 45 mins and get $20 gift card. Express your interest here: forms.gle/CtoZTwCovTv8....

AI research often aims to generate Wikipedia-like content, but Wikipedia’s success lies in the underlying processes and infrastructures that enable human collaboration. Rather than just mimicking content, we should use AI to support human collaboration. Our Wikibench paper is one such example:

"When a complex system is far from equilibrium, small islands of coherence in a sea of chaos have the capacity to shift the entire system to a higher order." Iya Prigogine, Noble prize-winning chemist

Check out our latest work on steerable alignment! We improve on a typical in-context learning approach with semantic retrieval by proposing metrics that instead select examples that best exemplify a group's shared values, enabling inference-time alignment to a particular group. ➡️

In-context learning can be an effective way to conduct value alignment of LLMs through examples, but when there are multiple pluralistic groups, are the best examples for one group also the ones for another? We explore this in our paper 🌟SPICA🌟 (🧵1/9)