🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

TRAFFIC CAM PHOTOBOOTH JUST WON A WEBBY!!!! 1 Person 4 Days $0 Budget 500,000 Users 800,000 Photos Taken 5 TV Appearances 1 Cease & Desist And Now 1 Webby

Lately, I've been seriously exploring what it could mean to move beyond the benchmarking paradigm in ML evaluation and it's led to some stats-y papers: (1) a critique of current experimental evals of prediction based interventions & (2) a framework for adverse events reporting.

Why RLHF stopped us from getting r1/o1 sooner and why we elected Trump/Biden/[Politician You Dislike]: We can use elections 🇺🇸🇺🇸🇺🇸 to understand why RLHF naturally suppresses reasoning (even if you fix the whole "RLHF isn't really RL" thing)

In LLM land, a slow model is annoying. In robotics, a slow model can be disastrous! Visible pauses at best, dangerously jerky motions at worst. But large VLAs are slow by nature. What can we do about this? An in-depth 🧵:

RLHF fine-tunes to a “mythical user” via aggregated feedback—but what if that user represents no one? Excited to share a new paper with Paul Gölz and Kunhe Yang “Distortion of AI Alignment: Does Preference Optimization Optimize for Preferences?” #AIAlignment #PluralisticAI #LLMs

Newest ⚡ reboot ⚡ 🎙️post: jessica dai and I discuss forecasting, and how people present unhelpful narratives about the future (mostly by picking on AI 2027, sorry guys) Why we should view the future as constructed, not predicted

just had 500 magazines delivered to my door :)

What more could we understand about the fractal, “jagged” edges of AI system deployments if we had better ways to listen to the people who interact with them? What a joy to work w jessica dai using individual experiences to inform AI evaluation (blog/ICML/arXiv links 👇)

we're launching a new issue of Kernel Magazine in two weeks!! join us for a launch party in SF @ Gray Area on July 17!

There's been an idea floating around policy spaces for some time now (eg. UN AI Advisory report, California Frontier AI report, etc) around the need for "AI monitoring", "AI incidents", "AI adverse reporting" etc. Jess is now seriously thinking about how to operationalize this:

Being a grad student is peak human existence. The end goal of all political and technological progress should be allowing everyone to be a grad student forever