Democratizing AI-Assisted Access to Knowledge! The Stanford OVAL Lab is leading an initiative to create a public AI Assistant that democratizes access to the world's knowledge. Our pilot program, already embraced by over 400,000 users, generates Wikipedia-like articles through

new paper! 🫡 we introduce 🪓AxBench, a scalable benchmark that evaluates interpretability techniques on two axes: concept detection and model steering. we find that: 🥇prompting and finetuning are still best 🥈supervised interp methods are effective 😮SAEs lag behind

🎉 For the first time ever: Collaborate with AI agents in real-time! Collaborative Gym UI is now IRB-approved and alive at cogym.saltlab.stanford.edu! A group of agents is eager to work with you. By providing feedback, you will see the agent's identity and its feedback to you!

Congratulations to Stanford NLP Group founder Christopher Manning for being elected to The National Academy of Engineering (NAE, National Academies) Class of 2025 for the development and dissemination of natural language processing methods.

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.

Come check out GenieWorksheets for programmable trustworthy LLM-based conversational agents and don’t miss the chance to talk to Harshit Joshi !

flying to Vienna 🇦🇹 for ACL to present Genie Worksheets (Monday 11am)! come and say hi if you want to talk about how to create controllable and reliable application layers on top of LLMs, knowledge discovery and curation, or just wanna hang

This is a great work to checkout! Cursor / Claude Code often attempts to accomplish tasks with shortcuts, such as just skipping a pytest. Research like this is increasingly necessary to build trustworthy and truly instruction-following systems.

.Stanford NLP Group has grown from a single office to our own wing! And our new lounge is coming together. But, more to get in place before the big reveal.

New paper! We explore a radical paradigm for AI evals: assessing LLMs on *unsolved* questions. Instead of contrived exams where progress ≠ value, we eval LLMs on organic, unsolved problems via reference-free LLM validation & community verification. LLMs solved ~10/500 so far:

Come and talk to Aryaman at COLM!

#COLM2025 Please drop by our ReasonIR poster at Poster3 #967 (11:00am - 1:00pm Wed) by Varsha Kishore 🥰 Happy to answer questions or chat online--feel free to DM! I've been exploring deep research training lately to empower reasoning+search for complex tasks💪 Stay tuned!

Many inconsistencies in Wikipedia discovered with the help of LLMs!

Had such a great time at the Stanford NLP Group reunion this past weekend, catching up with old friends and meeting new ones. Huge thanks to student organizers Dilara Soylu and Shicheng Liu for making everything seamless.

I'm recruiting multiple PhD students for Fall 2026 in Computer Science at JHU Computer Science 🍂 Apply to work on AI for social sciences/human behavior, social NLP, and LLMs for real-world applied domains you're passionate about! Learn more kristinagligoric.com & help spread the word!

Alex is a fantastic researcher! It would be a great working with him!