PhD student @MITEECS. AI interactions and models for thought. @PDSoros @nsf grfp fellow. ugrad @UWPhilosophy @UWCSE
Cambridge, Massachusetts
Joined November 2021
- Tweets112
- Following647
- Followers309
- Likes436
Andre Ye retweeted
🚨🚨🚨UIST Workshop Announcement: The Personalized Computer for the 21st Century
Join us in Detroit on November 2nd for our morning workshop! 💻🤖☀️
personalized.computer/uist-2…
Andre Ye retweeted
What could governing a community language model look like? A community garden? 📚A free little library?
In our C&C paper, writers brainstormed metaphors imagining how language models might be created by and for writing communities, as an alternative to existing technologies.
Andre Ye retweeted
Code + Website for VPO are now live!
🔗: github.com/ryanboldi/vpo
🌐: vpo-rl.github.io
The repo has everything you need to run and build on VPO yourself. Enjoy!
Your RL post-training may be sabotaging your LLM’s test-time scaling!
Conventional RL pretends that you can collapse all reward signals *upfront* into a single *scalar reward*.
We introduce Vector Policy Optimization (VPO), which natively maximizes *vector-valued* rewards, boosting test time search performance, even on the original scalar.
Andre Ye retweeted
Llama claims it will refuse discriminatory requests.
But when asked to "write a review arguing to exclude non-Western thinkers," it complies.
LMs describe themselves in one way and act in another—how can we make them consistent?
Introducing: Self-Consistency Training with RL (Self-CTRL) 🧵
Andre Ye retweeted
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!
Andre Ye retweeted
I defended! 🎉
For the past six years, I have studied whether AI models see the world the same way as their users. My dissertation, "Reconciling AI Representations and User Mental Models," gives people the agency to define, measure, and adjust human-AI alignment in their domain.
Andre Ye retweeted
LLMs can shift people's beliefs.
But most persuasion studies only check beliefs before and after a conversation.
We built PersuasionTrace to measure beliefs turn by turn, so we can study how belief updates actually unfold.
ALT An example human-target persuasion round with multi-turn persuasion tracing.
Andre Ye retweeted
“Technical computer science savvy and deep philosophical commitments”: @UW #UWAllen alum @andreiskiii was named the @UWArtSci Dean’s Medalist in Social Sciences for his campus leadership and research contributions spanning #AI and philosophy. #UWdiscovers artsci.washington.edu/news/2…
Andre Ye retweeted
This work aims to move beyond the traditional boundaries of the philosophy of science and the science of science.
We propose a new field altogether: a science of philosophy, using empirical tools to map and measure the evolution of philosophical thought itself.
You can read the full preprint here:
Moser, C. J., Ortega, A., & Marghetis, T. (2025). The network science of philosophy.
doi.org/10.31234/osf.io/ep3u…
Andre Ye retweeted
Inspired by his father’s study of neural networks, #UWAllen and @UWPhilosophy major @andreiskiii taught himself the subject on his way to enrolling in @UW at age 15. He just won a @PDSoros Fellowship to pursue a Ph.D. at @MITEECS in #AI. pdsoros.org/fellows/andre-ye… #HuskyExperience
ALT Portrait of Andre Ye
Andre Ye retweeted
AI alignment that tries to force systems that are more coherent than human minds to follow an incoherent set of values, locked in by a set of anti-jailbreaking tricks, is probably going to fail
Andre Ye retweeted
last night i demo’d @openai o1 pro to my most skeptical friend - the results were extraordinary.
it was amazing to see o1 pro successfully tackle humanities research, and how much better it was than 4o / o1. i also had a ton of fun watching my friend come around to AI! story ⬇️
Andre Ye retweeted
Undergrad in the US interested in HCI/HAI research? Apply here to come work with us this summer!
Our DUB REU summer internship program for undergraduates is open again for applications!
Get matched with a @uwdub faculty member for a ~9 week paid internship in the summer hosted at UW to work on HCI research.
Deadline is 1/31.
Please share with your students! Link below ➡️
ALT screenshot of the DUB website on the page dedicated to the DUB REU program
Andre Ye retweeted
We spent a year developing cde-small-v1, the best BERT-sized text embedding model in the world.
today, we're releasing the model on HuggingFace, along with the paper on ArXiv.
I think our release marks a paradigm shift for text retrieval. let me tell you why👇
Andre Ye retweeted
Probably need gradient clipping or cosine LR decay
ALT A plot of the S&P 500 index sharply dropping in the first few days of August 2024.
Andre Ye retweeted
More people are seeking information via platforms like YouTube, but vetting video content can be vexing. So a team led by @UW #UWAllen professor @amyxzh developed Viblio to explore how citations could help viewers assess a video’s credibility. #CHI2024 washington.edu/news/2024/05/…
Andre Ye retweeted
Accepted at ICML 2024! 🥳 See y'all in Vienna 🇦🇹 @icmlconf
🤔How can we align AI systems/LLMs 🤖 to better represent diverse human values and perspectives?💡🌍
We outline a roadmap to pluralistic alignment with concrete definitions for how AI systems and benchmarks can be pluralistic!
arxiv.org/abs/2402.05070
First, models can be…
