😅 is anyone else getting sick of chatbots? I wrote down a few thoughts on why chatbots are not the future of interfaces and how we can be more thoughtful: wattenberger.com/thoughts/boo-c…

Everyday I wake up feeling grateful and lucky that I am part of Delta Lab NU

How do existing knowledge, technologies, and power structures reinforce each other in Natural Language Processing research? We find that involvement in NLP research becomes more unequal as we move from unlabeled data to deployed systems. williamheld.com/coloniality

I’m excited I’ll be joining MIT next fall, for a shared interdisciplinary faculty position between Music (MIT Music & Theater Arts MIT School of Humanities, Arts and Social Sciences) & EECS. I’m recruiting PhD students MIT EECS for Fall 2024 (apply by Dec 15), and also Postdocs. See czhuang.github.io for details. Come join us!

Anna is such a thoughtful and talented researcher in Music, AI, and Co-Creation! Become one of her students!

Aligned LLMs should be helpful, harmless, and adopt user preferences. But whose preferences are we aligning to and what are unintended effects on global representation? We find SFT and Preference Tuning steer LLMs towards US English use and opinions. arxiv.org/abs/2402.15018 🧵

Before taking it out on your roommate for leaving dirty dishes out, you probably want to practice your conflict resolution skills first. Expert conflict resolution trainers, however, are EXPENSIVE. What if we practiced with an angry LLM instead? 😈 #CHI24 arxiv.org/abs/2309.12309

Thank you AK for sharing our Design2Code paper! Here’s my version of the story: To assess whether multimodal LLMs are ready to automate front-end engineering, we: - formalize the task of converting visual designs into code implementations - manually curate the Design2Code

✨Excited to announce our new paper on adaptive interventions for behavioral health change 🏃‍♂️, to appear at CHIL 2024 🏙️, now available on arxiv! arxiv.org/abs/2311.09483

We’re looking forward to ACL 2025 #ACL2024! From Diyi Yang’s group: • How Johnny Can Persuade LLMs to Jailbreak Them • Feedback Generation for Empowering Peer Counselors • Silent Signals, Loud Impact: Coded Dog Whistles • Impacts of LLM Alignment on Global Representation

AI has the potential to transform real-world domains. But can AI actually improve outcomes in live interactions? We conducted the first large-scale intervention of a Human-AI Approach that has statistically significant positive learning gains w/ 900 tutors & 1,800 K12 students.

Promoting my lab mates' work on establishing an evaluation for vision-language models for a really cool task: allowing designers to iterate quickly from low-fi wireframes to high-fidelity UI implementations. Cool mix of human-LLM interaction & dynamic automatic evals!

Todo lists, docs, email style – if you've got individual or team knowledge you want ChatGPT/Claude to have access to, Knoll (knollapp.com) is a personal RAG store from Stanford University that you can add any knowledge into. Instead of copy-pasting into your prompt every time,