Pinned
At the #Neurips2025 mechanistic interpretability workshop I gave a brief talk about Venetian glassmaking, since I think we face a similar moment in AI research today.
Here is a blog post summarizing the talk:
davidbau.com/archives/2025/…
David Bau
1,193 posts
David Bau
@davidbau
Computer Science Professor at Northeastern, Ex-Googler. Believes AI should be transparent. @[email protected] @davidbau.bsky.social baulab.info
- David's Tips on How to Read Pytorch github.com/davidbau/how-t… My quick, visual, principled introduction to pytorch. For teaching at MIT: five key concepts, five hackable Colab notebooks illustrating core deep net tools: GPU Tensors, Autograd, Optimizers, Modules, Dataloading.
- I want to show the NSF there would be broad support+utility for a "National Deep Inference" service for >100b LLMs. If your research would be enabled by an inference service on open LLMs w API access+overrides to internal activations, params, gradients: Please Like this thread!
- Dear MAGA friends, I have been worrying about STEM in the US a lot, because right now the Senate is writing new laws that cut 75% of the STEM budget in the US. Sorry for the long post, but the issue is really important, and I want to share what I know about it. The entire
- Help me out: what was the most important machine learning paper in 2024? I want to update my Famous Deep Learning Papers list (that I use in teaching) with new work from 2023/2024. papers.baulab.info Which new paper should I add?
- Why is interpretability the key to dominance in AI? Not winning the scaling race, or banning China. Our answer to OSTP/NSF, w/ Goodfire's @banburismus_ Transluce's @cogconfluence MIT's @dhadfieldmenell resilience.baulab.info/docs/AI_Action… Here's why:🧵 ↘️
- What does one neuron do in a deep net? Removing units reveals structure in classifiers and generators. Proceedings of the National Academy of Sciences dissect.csail.mit.edu github.com/davidbau/disse… pnas.org/content/pnas/e… @JunYanz @hen_str @AgataLapedriza @zhoubolei torralba
New NSF Grant to Establish National Deep Inference Fabric, Enhancing US Research on AI Transparency ow.ly/rrnn50Rw8N1 @NSF #HPC #TCIwire- DeepSeek R1 shows how important it is to be studying the internals of reasoning models. Try our code: Here @can_rager shows a method for auditing AI bias by probing the internal monologue. dsthoughts.baulab.info I'd be interested in your thoughts.
- PhD Applicants: remember that the Northeastern Computer Science PhD application deadline is Dec 15. It's a terrific time to do a PhD, with so many interesting things happening in AI. Apply here:
- Can the rules in a deep network be directly rewritten? Rewriting a Deep Generative Network (ECCV oral) rewriting.csail.mit.edu github.com/davidbau/rewri… news.mit.edu/2020/rewriting… Idea: Edit a deep network's weights by locating and changing one rule. Like rewriting a memory.
- When setting up my AI lab I faced a choice between Toronto and Boston. I chose Boston, my home and the world's best incubator for research talent. Here you can take a short stroll to meet with top minds in hundreds of fields from AI to astronomy, batteries to biotech.
- The NSF has invited us to propose an $18m (RI-1) project for infrastructure to help **you** (academic researchers) to study large language models. What should we build to help your LLM research? Respond with your thoughts. (Or answer 3Q's on a form forms.gle/s7EDW4V12mR7U3…) 🧵
- I love this summary of "what to learn in a PhD." What Jack points out seems like a simple principle, but living through the intellectual uncertainty and chaos of being lost in your research for a long time—it's emotionally much harder learn to do it than it sounds.the most satisfying takeaway from a phd is that you can solve problems far beyond your capabilities if you're willing to throw yourself at them again and again, over a long period, while staying open to new ideas it's not really about being smart. just curious & persistent
