2 days ago
MIT Open Scholarship is a completely revamped and significantly upgraded version of MIT’s institutional repository, also known as DSpace@MIT. For more than two decades, DSpace@MIT has served as the premier digital home for the Institute’s research, housing theses, working papers, technical reports, published articles, and more.
The new platform introduces a modernized, mobile-friendly user interface, improved search functionalities, and better capacity to show connections between researchers, organizations, and diverse scholarly outputs. Its new name reflects a broader commitment to the open sharing of knowledge, one that captures the true mission of the repository.
Huh. Glad they updated it, but the old name was cute. Ah, well.
7 days ago
Let me know if u guys want a video on how you (yes, you!) can be fake bi!!
13 Jul 26
Avid programmer that loves to push the limits of JavaScript
Dead blog of a good colleague.
28 May 26
We have some difficult news to share: Clover will be closing our doors this week. Thursday, May 28th will be our final day of restaurant service and meal box delivery.
RIP, Aaliyah. RIP…
27 May 26
This thesis argues that digital services need not come at the cost of our privacy. As a step towards this goal, this thesis introduces cryptographic protocols and secure systems that let users fetch and search over public data, without revealing anything about what they are fetching or searching for. We first present SimplePIR, an encryption scheme that can perform linear computations on encrypted data at near-unencrypted speed. Building on SimplePIR, we design Tiptoe, the first cryptographically-private web search engine. Tiptoe allows a user to search over hundreds of millions of webpages in seconds, all while revealing nothing about its query to the search provider; its core private-search techniques have seen adoption in Apple’s iOS (“Enhanced Visual Search”). Finally, we describe new private-information-retrieval protocols that circumvent Tiptoe’s linear-time bottleneck and could let private-search systems scale larger yet.
Coding agents can now produce code faster than humans can review it. Techniques that minimize the amount of code that can affect system correctness are therefore a major goal in the field. The Axon project used a coding agent to build a complete verified compiler — source language, operational semantics, optimizations, machine model, and mechanically checked correctness proofs — in 34 days under one supervisor.
The talk discusses how the project combined four validation techniques — testing, credible compilation, formal verification, and code audits - to maximize development velocity while minimizing the amount of unverified/unchecked code.
We present the overall architecture of the compiler, the roles that the different validation techniques played in the development, results from the Axon implementation, and implications for other agent centric development efforts.
see: https://arxiv.org/abs/2605.01660
19 Apr 26
MIT philosopher Sam Berstler provides analyzes the social dynamics accompanying open secrets. In many cases, she proposes, ignoring them is fine — but they may still have corrosive effects.
see: https://doi.org/10.1215/00318108-11676088
via: https://news.mit.edu/2026/why-bother-plausible-deniability-0417
In an article exploring the concept of plausible deniability, MIT philosopher Sam Berstler argues that our conversations cannot be understood simply by analyzing the words we use. Plausible deniability is bound up with other social practices that incentivize us to not be fully transparent.
BERSTLER: People who buy into the rhetoric of transparency can be setting back their own interests. Maybe speaking transparently is morally virtuous in some respects, but given the reality of our speech practices, transparency is not necessarily going to be the most effective way of handling things.
This is a huge discount against norms like radical honestly.
via: https://doi.org/10.1093/mind/fzaf016
07 Dec 25
music21is a Python-based toolkit for computer-aided musicology.
via: https://www.madmusicalscience.com/arcaexplanation.html via: https://dmitri.mycpanel.princeton.edu/index.html via: https://golem.ph.utexas.edu/category/2025/11/beyond_the_geometry_of_music.html
30 Nov 25
Most rejected, and all institutions let the clock run out on the deal. Guess Kornbluth set a trend. Trump has expanded the compact to all colleges: we’ll see if anyone bites.
23 Nov 25
the urge to try to attain a true “infinite rabbithole” of interpretation - a game whose secrets may never be fully plumbed - is a damn tempting one. and depth may be plumbed from even fixed sources!
Useful tools for navigating Moira lists at MIT
21 Nov 25
Machine Learning Systems provides a systematic framework for understanding and engineering machine learning (ML) systems. This textbook bridges the gap between theoretical foundations and practical engineering, emphasizing the systems perspective required to build effective AI solutions. Unlike resources that focus primarily on algorithms and model architectures, this book highlights the broader context in which ML systems operate, including data engineering, model optimization, hardware-aware training, and inference acceleration. Readers will develop the ability to reason about ML system architectures and apply enduring engineering principles for building flexible, efficient, and robust machine learning systems.
15 Nov 25
I love this article. It’s really persuasive.
09 Oct 25
We introduce a versatile method for finding prime numbers that display surprisingly intricate visual patterns— hypothetically, any desired pattern is possible, with only mild distortion. We use this method to locate several examples of large prime numbers that are, in and of themselves, self-referential works of art.
30 Sep 25
27 Sep 25
Good presentation on ologs.
Good presentation on ologs.
23 Sep 25
More than 20 percent of undergraduates have indicated sexual interest in another undergrad on a web app designed to decrease the effort required to begin a sexual relationship.
13 Aug 25
I’m Alice, a technical AI safety writer. I write the ML Safety Newsletter and my personal writing is on LessWrong. I have a background in technical ML, but pivoted to communications because I think this is where I can do the most good.