Angelina Wang

angelinawang.bsky.social

13 December 2025 21:31

Most LLM evals use API calls or offline inference, testing models in a memory-less silo. Our new Patterns paper shows this misses how LLMs actually behave in real user interfaces, where personalization and interaction history shape responses: arxiv.org/abs/2509.19364

Ece Takmaz

ecekt.bsky.social

27 October 2025 21:58

I felt very much at home at #ICCV2025! Here is the paper: arxiv.org/abs/2509.01453

Ece Takmaz

ecekt.bsky.social

27 October 2025 21:58

Just got back from Hawaii, where I presented a workshop paper on image memorability at @iccv.bsky.social 🌺 Coming from multimodal NLP, it was my first time attending a CV conference. Everywhere I looked, there were talks and posters that were incredibly interesting!

Jenn Wortman Vaughan

jennwv.bsky.social

25 October 2025 07:46

Looking for news on @msftresearch.bsky.social FATE internships & postdocs? YES, we will have positions, but bear with us — we expect postings to go up in mid-November and I will post an update here as soon as they do.

Dustin Wright

dustinbwright.com

15 October 2025 18:17

Which, whose, and how much knowledge do LLMs represent? I'm excited to share our preprint answering these questions: "Epistemic Diversity and Knowledge Collapse in Large Language Models" 📄Paper: arxiv.org/pdf/2510.04226 💻Code: github.com/dwright37/ll... 1/10