📢Do you work with or are you curious of computer using LLM agents? 👀
Our Notre Dame research team is studying how people evaluate LLM agents and we’d love your help!
⏱️ 90 min Zoom study
💵 $45 gift card
Sign up here if you’re interested: https://t.co/fU5HyqkFaM
Excited to report on another year of a strong showing of @NotreDame HCI community at #CHI2026. This milestone reflects the incredible hard work and creativity of our students, collaborators, and colleagues
https://t.co/cmsAb5MHZ0
Excited to share our paper, "Red Teaming LLMs as Socio-Technical Practice: From Exploration and Data Creation to Evaluation," received an Honorable Mention at #CHI2026!🥇
https://t.co/t9n0UDQy0d
We study how AI practitioners create, develop, and evaluate red teaming LLMs datasets
At @NMNH, an interactive exhibit used AI-generated personas to help kids explore real-world privacy choices—developed by @NotreDame & @JohnsHopkins researchers as part of an @NSF-funded project.
🔗 https://t.co/2VuGB9qE9m
@JHUCompSci@ND_CSE
I'll be presenting our 🏆 #CHI2025 paper on using Variation Theory of human learning and Structural Alignment Theory of sensemaking for co-adaptive machine teaching.
@TobyJLi@roboticwrestler
I’ll present our #CHI2025 paper, Hashtag Re-Appropriation for Audience Control on Recommendation-Driven Social Media Xiaohongshu (rednote).
📅 Tuesday, April 29
🕚 11:58 AM – 12:10 PM
📍 Room G401
If you're a rednote user, come come;
if you're not (yet!), also come come!
We find that aligning model training with human cognitive processes improves both the human's and model's understanding. If you're working on (or curious about!) interactive ML, human-AI alignment—let’s chat! 🌸
paper: https://t.co/4R6sICGWP9
video: https://t.co/4mvFf91Z8p
I'll be presenting our 🏆 #CHI2025 paper on using Variation Theory of human learning and Structural Alignment Theory of sensemaking for co-adaptive machine teaching.
@TobyJLi@roboticwrestler
Headed to #CHI2025? Don’t miss the exciting contributions, including 11 full papers and more, from Notre Dame researchers @ND_CSE@lucy_institute — and be sure to stop by and say hi!
https://t.co/d8hVNb0FmJ
We studied how GUI agents like OpenAI Operator and Claude behave under privacy & security threats.
We find:
❌ Fine-Print Injection worked up to 74% of the time.
❌ Humans performed even worse than the agents.
Here’s a 30-second flash video of what we found 👇
#LLM#privacy#AI
I'm here at #IUI2025! Today, I'll be giving a talk on the Limitations of the LLM-as-a-Judge Approach in Evaluating LLM Outputs in Expert Knowledge Tasks - co-led with @NoahZiems and co-authored with @TobyJLi, @Meng_CS, and @rametoyer. Come check it out!
https://t.co/GOMy2KQaTs
Each year, IUI acknowledges one past IUI paper that had the most impact and highest visibility since being published. This year IUI recognizes "Interactive Machine Learning" and hosted an engaging panel.
#IUI2025#HCI#AI
🎉 Meet our amazing team of student volunteers at IUI2025! 🎉
SVs, coming from diverse academic backgrounds, play a crucial role in making our conference a success!
A huge thank you to each and every one of them for their incredible dedication and effort! 🙌
#IUI2025#AI#HCI
Midwest Speech and Language Days will be held Apr 15-16 at @NotreDame! Abstract submissions are due Mar 20, and registration deadline is Mar 27. Financial assistance for students (lodging, poster printing) is available. https://t.co/1fFExngrE3