Ph.D. candidate in UMN Human AI Interaction in Real World! #XR #AI @UMN @ZJU_China
Minneapolis, MN
Joined September 2022
- Tweets215
- Following630
- Followers238
- Likes931
Pinned Tweet
Paper out at #UIST2026! We introduce Generative Proxy — a step toward spec-driven, #agentic engineering for heterogeneous #XR and #AR systems.
Amazing team: @CartonLiu, William Chastek, Eric Gonzalez(@_ejgonz_), Mar Gonzalez-Franco(@twi_mar) and Zhu-Tian Chen(@ZhutianChen)!!!
Heading to Detroit! will present
1. Paper (Generative Proxy) Tue 12:00pm-12:12pm (Nov 3) at Columbus, session: Mediated Reality and Adaptive XR Interfaces
2. Demo (Generative Proxy) Tue 6pm -9pm (Nov 3) at demo 1069, Ontario Zone 2
Also attend workshop "Revibing HCI" #UIST2026
Paper out at #UIST2026! We introduce Generative Proxy — a step toward spec-driven, #agentic engineering for heterogeneous #XR and #AR systems.
Amazing team: @CartonLiu, William Chastek, Eric Gonzalez(@_ejgonz_), Mar Gonzalez-Franco(@twi_mar) and Zhu-Tian Chen(@ZhutianChen)!!!
basically another type of AI reliance, but maybe in a romantic way lol
Do you have an AI confession?
This year, I collected 400+ handwritten secrets from strangers about their relationship with AI.
Check out the archive here.
WWW.WHATWETELLAI.COM
Designed by my brilliant friend, designer, and collaborator @arin_pantja
Preprint: arxiv.org/abs/2608.08971
Video: youtu.be/N1YcwAxcPaQ
#SmartAIGlasses? #XRHeadsets? #SpatialComputers? Nowadays #XR devices come in varying form factors, sensing, display, and interaction capabilities.
Why should we bother to adapt, redesign and rebuild the same UIs every time the device, scene, or user intent changes?
We envision a future where #XR interfaces transform smoothly across devices, scenes and intents — powered by #AI and #agentic tools!
Preprint: arxiv.org/abs/2608.08971 !!!
Congrats on @gloriakaku and @zhipinghci!
We got 2/2 COLM papers accepted! :)
Congratulations to @gloriakaku on building a dataset with nuanced human contextual privacy preferences and studying how LLMs fail to adapt to individual privacy preferences.
Congratulations to @zhipinghci on her work studying what agent autonomy means for privacy concerns and trust in agents. We found that when agents delegate control only upon detected risks, they achieve better perceived control—even more than when users are asked to confirm at every step.
Proud of both works, and grateful that the COLM reviewers recognized their value and made space for human-centered research at an LLM conference.
Time to chill and read books in the adorable spring!
Reflecting on my sense of #Body in this embodied world.
Plz bomb me with books/materials about human cognition!
Especially about space and embodiment.
medium.com/p/sense-of-body-i…
Catch me up and let’s connect at #CHI2026! Flying to the sky now✈️
Fun fact is I will show up in the
Same room P1 - Room 128 a lot!🤩
Wed for full paper
Mon for workshop
#XR + #AR + #AI in #HCI
ACM #CHI2026 Paper out! 🎉
Your #AI may cheat, and you may inappropriately rely on it in the future of #spatialcomputing era🚨!
Will humans team with AI to make spatial decision-making correctly under different #AR #XR visualizations?
We start exploring this research question by asking, “Can AR Embedded Visualizations Foster Appropriate Reliance on AI in Spatial Decision-Making? ”
arxiv: arxiv.org/abs/2507.14316
PS: I am actively seeking a PhD research summer internship starting from May 2026. Plz contact me!
Sincere shout-out for the whole team and great work: @KeithMaxwell99 @tongyunie @EvanSuma Victoria Interrante @ZhutianChen
Uncertainty will capture the subtle dynamics in human AI interaction, especially in physical world tasks. Fantastic work by @ChingYiTsai2 !!! Check it out!
As AI agents enter the physical world via smart glasses and robots, they need to interpret users’ ambiguous references, including imprecise pointing and underspecified language like “what’s that?” in complex real-world settings.
Interaction breakdowns occur when uncertainty is high, and errors are likely.
We introduce Uncertain Pointer at #CHI2026 to communicate system uncertainty with AR visualizations and facilitate user disambiguation!
Led by the amazing @ChingYiTsai2 with Nicole Tacconi and Andy Wilson 🧵👇
1/4
Xianhao Carton Liu ᯅ retweeted
🚀MIT Flow Matching and Diffusion Lecture 2026 Released (diffusion.csail.mit.edu/)!
We just released our new MIT 2026 course on flow matching and diffusion models! We teach the full stack of modern AI image, video, protein generators - theory and practice. We include:
📺 Videos: Step-by-step derivations.
📝 Notes: Mathematically self-contained lecture notes
💻 Coding: Hands-on exercises for every component
We fully improved last years’ iteration and added new topics: latent spaces, diffusion transformers, building language models with discrete diffusion models.
Everything is available here: diffusion.csail.mit.edu/
A huge thanks to Tommi Jaakkola for his support in making this class possible and Ashay Athalye (MIT SOUL) for the incredible production! Was fun to do this with @RShprints!
#MachineLearning #GenerativeAI #MIT #DiffusionModels #AI
Xianhao Carton Liu ᯅ retweeted
Assistant Professor Zhu-Tian Chen is exploring how people can naturally collaborate with AI in real-world spaces by designing augmented reality systems that make AI’s insights visible, tangible, and actionable in the physical world.
Read more: z.umn.edu/ChenVCStory
ALT VC Story
ALT VC Story
Xianhao Carton Liu ᯅ retweeted
Recent PhD graduate Yuxuan Huang is studying how VR can help facilitate wellbeing and stress reduction by immersing people in nature. He created multiple studies that compare different VR techniques and their ability to support relaxation.
Learn more: z.umn.edu/HuangAward
This is so fun lol as a climber
POV: Your AR glasses are your new climbing coach. 🧗
Duke senior Eleanor Taylor is using Spectacles to build the future of rock climbing. By combining hand tracking with personalized motion data, she’s making the wall more accessible for beginners.
The goal? A full suite of AR climbing games that know your height and wingspan better than you do.
Xianhao Carton Liu ᯅ retweeted
🤔 Can you tell if a user is overrelying on LLM by just observing how they type ⌨️, click ↖, and scroll🖱️on the UI?
📊 We took a data-driven approach & surfaced key behavioral indicators that might help mitigate overreliance on LLM.
👏 Led by @_ChangLiu @qinyizhou2024 @Frilk3
This Post is from an account that no longer exists. Learn more
Look forwards to Mar’s visit at UMN campus soon! @twi_mar
Join us, Will be an intellectual feast in #XR + #AI!
Our next colloquium is Monday, February 23rd, with Mar Gonzalez-Franco Google. The talk titled "Why do machines now show intelligence? And how can we use this to augment humans", will be held in Keller 3-180.
Learn more: z.umn.edu/GonzalezFrancoColl…
ACM #CHI2026 Paper out! 🎉
Your #AI may cheat, and you may inappropriately rely on it in the future of #spatialcomputing era🚨!
Will humans team with AI to make spatial decision-making correctly under different #AR #XR visualizations?
We start exploring this research question by asking, “Can AR Embedded Visualizations Foster Appropriate Reliance on AI in Spatial Decision-Making? ”
arxiv: arxiv.org/abs/2507.14316
PS: I am actively seeking a PhD research summer internship starting from May 2026. Plz contact me!
Sincere shout-out for the whole team and great work: @KeithMaxwell99 @tongyunie @EvanSuma Victoria Interrante @ZhutianChen
PS: I am actively seeking a PhD research summer internship starting from May 2026. Plz contact me!
Check my homepage for more details.
carton22.github.io/
