Ece Kamar

2/10 top papers this week by MSR AI Frontiers :)

Excited to share Momento from the AI Frontiers Lab at Microsoft Research! 🚀 This new approach makes reasoning models more efficient by teaching them how to manage their own memory. A great start to the meta-reasoning capabilities agents will need. 🧠🤖 x.com/DimitrisPapail…

Excited to share our new work with an amazing team at @MSFTResearch @yzeng58 @ShivamGarg91462 @ChenLingjiao @tanghao95 @ZiyanWang98 @AhmedHAwadallah @erichorvitz @JohnCLangford @DimitrisPapail

Reasoning models think hard — but all that thinking fills up your KV cache fast. Memento fixes this: the model compresses its own chain-of-thought mid-generation, flushing old KV entries after each block. 2-3× less peak KV cache, ~2× throughput — accuracy largely preserved. The cool part: deciding what to remember and what to forget is a capability the model acquires through training — not something you bolt on. Excited about where this goes — especially for agents.

More great work coming out of our AI Frontiers lab at Microsoft Research: Best in class CUA verifiers.

We believe open, small, and efficient models are the key to unlocking the future of Agentic AI. Can't wait to see what the community builds! 🚀 #AIFrontiers #SmallLanguageModels #Phi #AgenticAI #MSR

Phi-4-reasoning-vision and the lessons of training a multimodal reasoning model microsoft.com/en-us/research…

At the AI Frontiers Lab, we’re releasing the last member of our Phi family. #MSFTResearchhttps://www.microsoft.com/en-us/research/blog/phi-4-reasoning-vision-and-the-lessons-of-training-a-multimodal-reasoning-model/

In the fast race of AI, I’m pausing this Sunday @UW to reflect on my journey and share learnings with the next generation of Turkish women leaders. Grateful for those who paved the way. Let’s carry the flag forward! 🇹🇷✨ #InternationalWomensDay #WomenInAI #TACAWA

Had such a fun time chatting with Jordan Wilson on his Everyday AI podcast! We dug into why the future of AI isn’t just about bigger models—it’s about societies of smaller, collaborative agents and how this shift will reshape the way organizations build with AI. @MSFTResearch

Inside the Society of Agents: Why AI Teamwork Beats Bigger Models x.com/i/broadcasts/1…

Heading to Singapore to give a keynote at @RealAAAI on AI agents and the future they present for us. See you at AAAI!

Try FARA 7B on Hugging Face & Foundry, or experience it via the Magentic UI! 👇 Technical report: aka.ms/fara-techreport Hugging Face: huggingface.co/microsoft/fara… Microsoft Foundry: aka.ms/foundry-fara-7b Foundry Labs blog: labs.ai.azure.com/projects/fara-…

This builds on our successful #Phi family of models and pushes our SLM work firmly into the agentic future. Key benefits: Highly capable for complex tasks. Extremely efficient for #OnDeviceAI. Enables faster, more private user experiences.

Big news from #AIFrontiers @ #MicrosoftResearch! We're thrilled to release FARA 7B—an incredibly efficient Agentic SLM (7B parameters) for computer use! 🚀 microsoft.com/en-us/research…

Microsoft researchers are driving trends and tackling major challenges in the midst of the ever-changing AI era. Check out Trista Chen, Flavio Griggio, Ahmed Awadallah, Jina Suh, and Cecily Morrison's conversations with Forward Future, a daily AI newsletter, to learn more about the people building the systems that are becoming part of our daily lives. msft.it/6015tJvdx

Magentic Marketplace is live from AI Frontiers @MSFTResearch—an open simulation for studying agentic markets. @thenewstack dives into why we test in the open: agents will reshape work, so we must map risks & failure modes first. #AgenticAI #AIResponsibility #OpenSource

Magentic Marketplace is a simulation environment for agentic markets, a new project from @Microsoft to explore multi-agent collaboration. Feat. @ecekamar thenewstack.io/microsoft-laun…

Launching Magentic Marketplace from AI Frontiers. Agents are about to reshape how we work, trade, and collaborate. Our job is to understand the system—and its failure modes—before it scales. Thoughtful research first, deployment second. techcrunch.com/2025/11/05/mic… via @techcrunch