Google Research Scientist PhD Internship application for 2026. (For Europe) Due date: 24th Oct. https://t.co/tOm5VxFIvR I will be hiring interns on safety, security, and privacy in (generative/agentic) AI in Zurich.

I'm recruiting students for fall 2026 thru @LTIatCMU & @CMU_EPP, in: 1. Privacy & security of LLMs, coding, long horizon & embodied agents (robotics) 2. Tiny local llms 3. AI for scientific reasoning, esp. chemistry 4. Latent reasoning 5. anything YOU are passionate about!

Grateful and happy to receive the ACM CCS @acm_ccs Test of Time Award for our “Privacy-Preserving Deep Learning” paper with Vitaly Shmatikov back in 2015. First reaction: “10 years, man!”

غافل مَباش اَر عاقلی؛ دریاب اگر صاحب‌دلی باشد که نتوان یافتن دیگر چنین ایـــــــــام را صدای شجریان است از غزل سعدیِ جان آلبوم درخیال ؛ به صدای تارِ مجید درخشانی

Proud to announce an official Gold Medal at #IMO2025🥇 The IMO committee has certified the result from our general-purpose Gemini system—a landmark moment for our team and for the future of AI reasoning. https://t.co/QgbfHL5X15 (1/n) Highlights in thread:

Check out two recent blog posts from our team: 1) Graph Foundation Models, and how they help achieve 3-40x in precision: https://t.co/YkRQOfY1bj 2) Enabling efficient Multi-Vector retrieval, via MUVERA: https://t.co/nCvMssWtQR (based on our NeuRIPS'24 paper).

از کران تا به کران لشکر ظلم است ولی از ازل تا به ابد فرصت درویشان است

روز ملی دماوند، فخر فلات ایران، بر ایرانیان خجسته باد عکس‌ از مهدی نورمحمدی

چون اصل های بیخ شان از راه پنهان بشکنم

روزی دو باغ طاغیان گر سبز بینی غم مخور

‌ ‌ دل قوی دار که بنیاد بقا محکم از اوست حضرت #سعدی ‌ ‌

دیار عاشقی‌هایم… مردم ایران ❤️

Joined Google Research. Zurich.

Today, we celebrate International Women in Mathematics Day in memory of Maryam Mirzakhani, the first woman to win the Fields Medal (2014), to honor the contributions and achievements in women in mathematics in hopes to inspire future generations.

Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory

Leave-one-out Distinguishability in Machine Learning https://t.co/aY20JZe6WP, Low-Cost High-Power Membership Inference Attacks https://t.co/53q8nser0U, Minerva: A Programmable Memory Test Benchmark for Language Models https://t.co/wmg8xAVhAV,

Slides of my ICLR talk on "Memory, Information Flow, and Privacy": https://t.co/Qrqpr7TnWd I presented the underlying story of these papers: Rethinking Hallucinations: Correctness, Consistency, and Prompt Multiplicity https://t.co/HIfR0J77Tc,

OpenAI Operator enables users to automate complex tasks, e.g., travel plans. Services, e.g., Expedia, use chatbots. Soon, these two ends are going to communicate, forming agentic networks. What would these networks enable? what are their risks? and how to secure them? 🧵1/n

MSR Cambridge

🧙 I am recruiting PhD students and postdocs to work together on making sure AI Systems and Agents are built safe and respect privacy (+ other social values). Apply to UMass Amherst @manningcics and enjoy a beautiful town in Western Massachusetts. Reach out if you have questions!