blank

I’m doing a 30 day research sprint on linear attention.

I’m doing a 30 day research sprint on linear attention.

I’m doing a 30 day research sprint on linear attention.

I’m doing a 30 day research sprint on linear attention.

LLM inference workloads have been moving towards longer context lengths, stemming from usage from long-running agents and requiring working in large codebases/contexts.

The quest for a new white box modeling paradigm While we have witnessed empirical evidence of the success of deep learning, much of it is attributable to trial and error and not guided by underlying mathematical principles.

Deep learning is currently dominated by parametric models, which are models with a fixed number of parameters regardless of the size of the training dataset.

A Markov chain that is aperiodic and irreducible will eventually converge to a stationary distribution.

Rhythm is just a $200 controller and some hopefully-not-too-complicated open source software setup away!

In this article, I will show you how you can generate trackback requests to external websites to link back to your static site like Jekyll or Hugo.

\[\newcommand{\rcal}{\mathcal{R}} \newcommand{\lcal}{\mathcal{L}} \newcommand{\mcal}{\mathcal{M}} \newcommand{\mocal}{\overline{\mathcal{M}}} \newcommand{\mocalp}{\overline{\mathcal{M}}^\perp} \newcommand{\mcalp}{\mathcal{M}^\perp} \newcommand{\sse}{\subs…

If you’re curious about the infamous steam tunnels at CMU, or what the views from the roof of Wean Hall looks like, this is for you!

This semester (Spring 2023), I took 15-712 Advanced Operating Systems and Distributed Systems under professor Phil Gibbons and his TA and also PhD advisee Val Choung.