🥳📕Happy to share that our paper "Offline Evaluation of Set-Based Text-to-Image Generation" got accepted at SIGIR-AP 2025 This was the outcome of my internship at Google Brain W/ Fernando Diaz #SIGIR #SIGIRAP2024

i also want to point ppl to the survey prepared by Danny To Eun Kim, Alireza Salemi, Andrew Drozdov, Hamed Zamani, and i reviewing more opportunities for approaching rag from a retrieval perspective. enjoy! arxiv.org/abs/2407.12982

Very happy my intern work Google AI accepted to #NeurIPS2024 NeurIPS Conference We use Gaussian Process for user modeling to capture multiple interests in an elegant way. Grateful to my collaborators and mentors Ofer Meshi Masrour Zoghi, Fernando Diaz, Steve, Maryam Karimzadehgan, Craig Boutilier

Can LLMs cater to diverse cultures in text generation? We find: 1️⃣lexical variance across nationalities 2️⃣culturally salient words 3️⃣weak correlation w/ cultural values 📜arxiv.org/abs/2406.11565 🤗huggingface.co/datasets/shail… 💻github.com/shaily99/eecc 🎉EMNLP 2025🎉 w/ Fernando Diaz 🧵

🤖💡 Brainstorming with ChatGPT’s help? ⚠️ Be careful what you reveal about yourself, because large language models might give you stereotypical or biased recommendations. Preprint here: arxiv.org/abs/2410.05613 🧵1/7

Do models personalize results when we ask them to and avoid stereotypes otherwise? No. Well are they at least transparent about it? Also no… ⚠️ If the model can infer your race you might get racially biased recommendations! 📄 Preprint: arxiv.org/pdf/2410.05613 🧵1/8

New Fingerprint Mass Spectrometry Method Paves the Way to Solving the Proteome - caltech.edu caltech.edu/about/news/new…

What a coincidence that we released announcements about LLM bias on the same day! But our conclusions were different - OpenAI found minimal bias while we found significant bias. 👀 Why is this? 🧐 🧵

I'm on the job market for tt faculty positions! I conduct research to resist harmful tech, data and AI and empower communities to imagine and build better data and AI futures. Please let me know if there are any positions I might be a good fit for!

The AI Interdisciplinary Institute at Univ. of Maryland (AIM) is hiring 40 new faculty members in all areas of AI, particularly: - accessibility, - sustainability, - social justice, and - learning; building on computational, humanistic, or social scientific approaches to AI. >

Today we'll be presenting the Tutorial on Retrieval-Enhanced Machine Learning (REML). Come by to learn about the emerging design patterns in this space and see how retrieval can be used beyond RAG. In collaboration w/ the amazing Fernando Diaz Danny To Eun Kim (✈️SIGIR🇮🇹) Alireza Salemi Hamed Zamani

If you're interested in OpenAI including shopping results, you might also be interested in Danny To Eun Kim (@teknology.bsky.social)'s paper relating retrieval diversity/fairness and generation by downstream RAG models. This has implications for individuals selling products online. arxiv.org/abs/2409.11598

Ever trusted a metric that works great on average, only for it to fail in your specific use case? In our #NAACL2025 paper (w/ Fernando Diaz), we show why global evaluations are not enough and why context matters more than you think. 📄 aclanthology.org/2025.findings-… #NLP #Evaluation (🧵1/9)