Rob Nowak
@rdnowak
Followers
2K
Following
3K
Media
173
Statuses
1K
Director of the Center for the Advancement of Progress
Ho-Chunk land
Joined April 2009
Attended a talk today about the possibilitiy of AI systems having qualia, such as pain or pleasure. Reminded me of the great novel “Blindsight” by Peter Watts, that considered whether consciousness is necessary for intelligence. Maybe we already have our answer?
2
0
13
Please share widely. The UW–Madison ECE Department is recruiting at the assistant, associate, or full professor in foundational AI and machine learning
jobs.wisc.edu
Current Employees: If you are currently employed at any of the Universities of Wisconsin, log in to Workday to apply through the internal application process.Job Category:FacultyEmployment Type:Reg...
2
41
96
This is incredible: In a blistering dismantling of HHS Secretary RFK Jr., Sen. Warnock noted that we're seeing child deaths from measles for the 1st time in decades. "You are a hazard to the health of the American people." Make sure everyone sees this. https://t.co/ND1lGjyPUH
362
2K
7K
Congratulations, Jifan. Working with you has been super rewarding! I can't wait to see what you accomplish next.
Defended my PhD yesterday! I wrote a blogpost that distilled my research into a single takeaway (in 🧵) Thank you @rdnowak for the amazing four years. Now that I see it, he truly has the magic potion that helped so many of his students turn into great researchers.
0
0
21
Congratulations, Sharon!
It’s official: I got my tenure! Immensely grateful to my colleagues, students, friends, and family who have supported me on this journey. On, Wisconsin!
1
0
11
Wow. Amazing progress. Congrats to the whole team!
It was amazing to be part of this effort. Huge shout out to the team, and all the incredible pre-training and post-training efforts that ensure Gemini is the leading frontier model! https://t.co/KUHpY8y1qL
0
0
4
Looking forward to seeing folks tomorrow afternoon!
Working in active learning, I often hear a common complaint — active learning doesn’t work sometimes. Our ICML 2025 paper JUST WORKS* when you have class imbalance. @rdnowak will present it Tuesday afternoon. Short 🧵below Paper:
0
4
18
... although we don't know if Grok 4 is consulting with the boss to dominate HumorBench, https://t.co/X8pOMUOP9o :)
techcrunch.com
Elon Musk's newly launched AI chatbot, Grok 4, seemed to reference Musk's posts on social media before answering controversial questions.
Releasing HumorBench today. Grok 4 is🥇 on this uncontaminated, non-STEM humor reasoning benchmark. 🫡🫡@xai Here are couple things I find surprising👇 1. this benchmark yields an almost perfect rank correlation with ARC-AGI. Yet the task of reasoning about New Yorker style
1
0
4
Surprising to also see how STEM fine-tuning transfers to reasoning about humor and jokes!
Whoa... Grok 4 beats o3 on our never-released benchmark: HumorBench, a non-STEM reasoning benchmark that measures humor comprehension. The task is simple: given a New Yorker Caption Contest cartoon and caption, explain the joke.
0
0
5
LLM scaling, pretraining or test-time, has lost its magic when it comes to humor understanding. Despite advances from GPT-4 to o3, we've seen zero and sometimes negative progress on humor understanding of the New Yorker cartoons. 🧵Our new paper on how to unlock expert-level
1
6
15
🥲 Current image diffusion models struggle to generate complex images such as "a mustachioed squirrel holding an ax-shaped guitar on a stage." 😎 Find a proxy prompt where rare associations are replaced with common ones, and alternate between the two prompts during generation!
🚀our paper got in #ICLR2025! Rare-to-Frequent: Unlocking Compositional Generation Power of Diffusion Models on Rare Concepts with LLM Guidance TL;DR: Exposing frequent concepts using LLMs can boost the compositionality of diffusion models—boosting #FLUX,#SD3 using #GPT! 🧵👇
2
16
67
elated to announce that I also joined academic twitter 2016
1
0
8
Wow! Awesome. Thanks so much to the authors for pulling this valuable information together in a beautiful and accessible form.
This new textbook by @ml_angelopoulos Rina Barber and @stats_stephen looks really neat!
1
0
17
Laura is an amazing researcher. Excited about this talk!
The UW-Madison ECE is thrilled to host Associate Professor Laura Balzano from the University of Michigan for an insightful talk on finding linear structure in modern machine learning data. For full event details: https://t.co/LtZiudOnDw
0
4
22
Looking forward to this talk!
I'm thrilled to announce that François Charton (@f_charton @AIatMeta) will be kicking off our new AI for Science seminar series next Wednesday. He is at the forefront of using AI for mathematics, cryptography, and theoretical physics. @datascience_uw
https://t.co/8wTojx76yt
0
0
6
“And which great efficiency expert will lead this important new effort to streamline?” “Actually there will be two people in charge”
313
5K
60K
Looks like an awesome workshop!
Two more weeks to submit your work on tensors/low-rank factorizations in the workshop
0
0
2
Two more weeks to submit your work on tensors/low-rank factorizations in the workshop
Less than two weeks to submit your papers on: 📈 #lowrank adapters and #factorizations 🧊 #tensor networks 🔌 probabilistic #circuits 🎓 #theory of factorizations to the first workshop on connecting them in #AI #ML at @RealAAAI please share! 🔁 👇👇👇 https://t.co/m65ScBfTyL
0
2
5