Michael Ryan
@michaelryan207
Incoming PhD Student @stanfordnlp || Working on DSPy 🧩 || Prev @GeorgiaTech @Microsoft @SnowflakeDB
New #ACL2025NLP Paper! 🎉 Curious what AI thinks about YOU? We interact with AI every day, offering all kinds of feedback, both implicit ✏️ and explicit 👍. What if we used this feedback to personalize your AI assistant to you? Introducing SynthesizeMe! An approach for…
.@stanfordnlp papers at @aclmeeting in Vienna next week: • HumT DumT: Measuring and controlling human-like language in LLMs @chengmyra1 @sunnyyuych @jurafsky • Controllable and Reliable Knowledge-Intensive Task Agents with Declarative GenieWorksheets @harshitj__ @ShichengGLiu…
The #SIGIR2025 Best Paper just awarded to the WARP engine for fast late interaction! Congrats to Luca Scheerer🎉 WARP was his @ETH_en MS thesis, completed while visiting us at @StanfordNLP. Incidentally, it's the fifth Paper Award for a ColBERT paper since 2020!* Luca did an…
📢 If you’re at #SIGIR2025 this week, make sure to be at Luca Scheerer’s paper talk: “WARP: An Efficient Engine for Multi-Vector Retrieval” (Wednesday 11am) WARP makes PLAID, the famous ludicrously fast ColBERT engine, another 3x faster on CPUs. With the usual ColBERT quality!
Are AI scientists already better than human researchers? We recruited 43 PhD students to spend 3 months executing research ideas proposed by an LLM agent vs human experts. Main finding: LLM ideas result in worse projects than human ideas.
AI companions aren’t science fiction anymore 🤖💬❤️ Thousands are turning to AI chatbots for emotional connection – finding comfort, sharing secrets, and even falling in love. But as AI companionship grows, the line between real and artificial relationships blurs. 📰 “Can A.I.…
Earlier we released SynthesizeMe #ACL2025NLP — a method for building a personalized persona prompt from your LLM interactions 🤖. Wanted to highlight this video from the thread that gives a lot of intuition for how SynthesizeMe (and related prompt opt algos) work! 👇
Given a small set of pairwise preferences, SynthesizeMe builds a personalized prompt in 3 steps: 1️⃣ Generate reasoning traces via LLM; keep only those that match ground truth (rejection sampling). 2️⃣ Synthesize a user persona from the validated traces. 3️⃣ Select the most…
Super comprehensive paper from @EchoShao8899 and Humiska asking a fundamental question: what tasks do people *want* automated by AI? This is my favorite figure, tons to learn from this work!
🚨 70 million US workers are about to face their biggest workplace transmission due to AI agents. But nobody asks them what they want. While AI races to automate everything, we took a different approach: auditing what workers want vs. what AI can do across the US workforce.🧵
super cool work from the goat @michaelryan207 on personalization for LM's. Personalization is a (maybe even 'the') core problem in how humans will interact with language models going forward. Michael is a (maybe even 'the') goat researcher + presenter. This is very cool work!
New #ACL2025NLP Paper! 🎉 Curious what AI thinks about YOU? We interact with AI every day, offering all kinds of feedback, both implicit ✏️ and explicit 👍. What if we used this feedback to personalize your AI assistant to you? Introducing SynthesizeMe! An approach for…
SynthesizeMe personalizes LLMs by inducing synthetic user personas from interaction reasoning, without relying on demographics or predefined categories 🫡 Try it yourself: try-synthesize.me
New #ACL2025NLP Paper! 🎉 Curious what AI thinks about YOU? We interact with AI every day, offering all kinds of feedback, both implicit ✏️ and explicit 👍. What if we used this feedback to personalize your AI assistant to you? Introducing SynthesizeMe! An approach for…
Turn yourself into a prompt with "SynthesizeMe" so any LLM can be personalized for you? 👀 Another best-in-class application of dspy.MIPROv2.
New #ACL2025NLP Paper! 🎉 Curious what AI thinks about YOU? We interact with AI every day, offering all kinds of feedback, both implicit ✏️ and explicit 👍. What if we used this feedback to personalize your AI assistant to you? Introducing SynthesizeMe! An approach for…
You can build prettyyyy good user models from just pairwise preferences too!!
New #ACL2025NLP Paper! 🎉 Curious what AI thinks about YOU? We interact with AI every day, offering all kinds of feedback, both implicit ✏️ and explicit 👍. What if we used this feedback to personalize your AI assistant to you? Introducing SynthesizeMe! An approach for…