Stanford Human-Computer Interaction Group

Updates about Stanford's HCI Group. Account run by Helena Vasconcelos. Visit for more!

[Stanford Human-Computer Interaction Group @StanfordHCI]

🎉🎉🎉🎊🎊🎊🎊🎊🥳🥳🥳🥳🥳🥳🥳🥳🥳🥳🥳🥳🥳🥳🥳 congratulations to our good friend and incredibly intelligent and sharp collaborator @michelle123lam

[Michelle Lam @michelle123lam]

After over 10 years at Stanford, it's time to leave :)

I will be joining UT Austin's CS department as an assistant professor in fall 2027! If you're excited to envision the future of interaction with AI, I'm recruiting PhD students this cycle. Come join me!

[Michelle Lam @michelle123lam]

If you're interested, I'd love to get to know you: https://t.co/SEJv3pUQq0

Thank you to my advisors @msbernst & @landay, without whom I would certainly not be on this path of HCI research. Thank you @StanfordHCI, my home for all these years. You all make it hard to leave!

[Michelle Lam @michelle123lam]

After over 10 years at Stanford, it's time to leave :)

I will be joining UT Austin's CS department as an assistant professor in fall 2027! If you're excited to envision the future of interaction with AI, I'm recruiting PhD students this cycle. Come join me!

[Omar Shaikh @oshaikh13]

We upgraded Tabracadabra 🎉 to bring an entire context-aware assistant (not just tab to autocomplete!) to any textbox. It's pretty great if you hate switching between the chat interface and what you're working on. We're also open-sourcing, so you can try it out!🧵

[Stanford Human-Computer Interaction Group @StanfordHCI]

👀

[MIT HCI @mithci]

We're back! The MIT HCI group has grown, and we couldn't be more excited. A huge welcome to our newest faculty (@mitchellgordon, @huangcza, @ZanaBucinca & @jas_x_flowers) & students joining @arvindsatya1, @karger, Stefanie Mueller, Rob Miller & Daniel Jackson. Give us a follow!

[Vishakh Padmakumar @vishakh_pk]

People are increasingly worried that AI tools make us overreliant.
But how do we actually measure this? We introduce Offloading Score, a measure of reliance based on the fraction of cognitive effort offloaded to AI while completing a task.
In a controlled user study, Offloading Score detects increased reliance under time pressure, while several common alternatives do not.
(1/9)

[Diyi Yang @Diyi_Yang]

We propose a new way to quantify AI overreliance: the Offloading Score 🧐 @vishakh_pk

It measures the fraction of cognitive work you hand off to AI 🤖 via simulating how you'd have done each step without AI, then counting the steps the AI saved. It works directly from interaction traces (keystrokes, screenshots), so it's reusable across many tools!!

[Myra Cheng @chengmyra1]

ELEPHANT (our social sycophancy benchmark) is at #ICLR2026!

+ our work on steering LLMs' assumptions is at the Re-Align workshop!

Poster↓

Can't be there IRL but pls reach out to chat about sycophancy, social NLP, LLMs to promote agency & well-being, anything else!

[Stanford Human-Computer Interaction Group @StanfordHCI]

👏👏👏👏👏

[Michael Y. Li @michaelyli_]

Can a language model learn, end-to-end, what to keep in its own KV cache and what to throw away? Can it learn to forget while it learns to reason?

Deep learning's central lesson: capability emerges from end-to-end optimization, not heuristics/strong inductive biases. But for efficiency, we rely heavily on hand-designed approaches.

🗑️ Introducing Neural Garbage Collection (NGC): we train a language model to jointly reason and manage its own KV cache, using reinforcement learning with outcome-based task reward alone. No SFT, no proxy objectives, no summarization in natural language.

New paper with @jubayer_hamid, Emily Fox, and @noahdgoodman!

[Michael Y. Li @michaelyli_]

Can a language model learn, end-to-end, what to keep in its own KV cache and what to throw away? Can it learn to forget while it learns to reason?

Deep learning's central lesson: capability emerges from end-to-end optimization, not heuristics/strong inductive biases. But for efficiency, we rely heavily on hand-designed approaches.

🗑️ Introducing Neural Garbage Collection (NGC): we train a language model to jointly reason and manage its own KV cache, using reinforcement learning with outcome-based task reward alone. No SFT, no proxy objectives, no summarization in natural language.

New paper with @jubayer_hamid, Emily Fox, and @noahdgoodman!

[Diyi Yang @Diyi_Yang]

What if LLMs could help humans be better at helping other humans? Our work leverages AI as a practice partner and mentor for building social skills, from counseling to conflict resolution @StanfordHAI. Here's how it works:

https://t.co/iBxutrWBZQ

[Omar Shaikh @oshaikh13]

very cool idea @OpenAI😅

I’m really excited about this research preview- learning from how people interact with their computer BEYOND chat will unlock a ton of really cool interactions

(also we have a paper on this!! https://t.co/N3Y1cT4aZj)

[seancrotty @seancrotty]

Enjoying a talk at Stanford by Mike Kuniavsky and Elizbeth Goodman entitled "Observing the User Experience" in the 2026 AI environment. They are joint authors of that book and are presenting as part of Stanford's Human-Computer Interaction (HCI) Seminar series. Nice to see the evolution of their work since Mike and I worked together at Xerox PARC.

https://t.co/6dWOMy9kxV

@Stanford @StanfordHCI #AI @mikekuniavsky @egoodman #UserExperience #UXResearch #UXR @PARCinc

[Omar Shaikh @oshaikh13]

This is one of my fav figures in our paper. You can:

1. Identify a user's objective by observing general interaction with their computer.

2. Use it to construct a "just in time" rubric.

3. Sample bunch from model and SCALE TEST TIME COMPUTE ON LITERALLY ANY OPEN-ENDED TASK?!?

[Hari Subramonyam @HariSubramonyam]

@zoekaputa will present our paper on Friday at #CHI2026:

SimStep: Human-in-the-Loop Authoring of Interactive Educational Simulations Through Task-Level Abstractions

📆 Fri, 17 Apr
🕚 11:15 AM–12:45 PM
📍 P1 - Room 117 (Convention Centre - P1)
A thread 🧵

[James Landay @landay]

Star co-advised student @michelle123lam giving a talk on very cool work this morning at #chi2026 in room 111 at noon. Come see it! Best paper honorable mention winner.

[Michelle Lam @michelle123lam]

Thanks so much to co-authors @oshaikh13, Hallie Xu, @azguo2, @Diyi_Yang, @jeffrey_heer, @landay, & @msbernst! We’re grateful that this work received a CHI best paper honorable mention!

paper: https://t.co/jRh5aDyTuM
waitlist: https://t.co/4PsdjtKpgp
talk: Th 4/16 12pm @ P1-111

[Michelle Lam @michelle123lam]

Most of what I actually need help with, I never think to tell a model. But why is it on me to remember?

Our new paper asks: what if AI could proactively specialize to individuals and the tasks they’re carrying out at this very moment? 🧵

[Shardul Sapkota @sapkotashardul_]

There’s much more to health behavior change than just step count and calories burned. Really excited to share our work on LLM-augmented behavior change interactions, led by @mjoerke!

[Stanford HAI @StanfordHAI]

Congratulations to @mjoerke, @landay, @EmmaBrunskill, and the rest of the team on receiving a Best Paper Award at #CHI2026! We're proud to support innovative research that bridges cutting-edge AI with human-centered design. 🎉 @acm_chi