2026-07-22

9 saved articles
BackLibrary
ArticleRead
Essence of linear algebra
YouTube
A free course offering the core concept of linear algebra with a visuals-first approach.
--
Models are worse at reviewing their own code
Rodrigo Caridad / Greptile
We noticed GPT reviewed Claude's code better than its own, and vice versa. So we ran a thousand buggy PRs through both models to find out why.
--
GitHub - plasma-ai/fractal: Hierarchical agent loops with recursive self-organization.
GitHub
Hierarchical agent loops with recursive self-organization. - plasma-ai/fractal
--
People use fast and flat simulation to reason about new games
Tenenbaum, Joshua B. / Nature
Games have long been a microcosm for studying planning and reasoning in both natural and artificial intelligence, often focusing on expert-level or even super-human play1–6. But real life also pushes human intelligence along a different frontier, requiring people to flexibly navigate decision-making problems that they have never thought about before. Here we use novice gameplay to study how people reason about new problem settings. Through a series of large-scale behavioural studies with over 1,
--
agentOS — Virtual operating system for agents
agentOS
A virtual operating system for agents, with filesystem, execution, and orchestration in one lightweight library.
--
vinext — The Next.js API surface, reimplemented on Vite
vinext.dev
Take any Next.js app and deploy it anywhere with one command. App Router, Pages Router, RSC, ISR — all on Vite.
--
Mage-Flow: An Efficient Native-Resolution Foundation Model for Image Generation and Editing
[Submitted on 21 Jul 2026] / arXiv.org
Large-scale visual generators are increasingly capable but costly to train, fine-tune, and deploy. We introduce Mage-Flow, a compact 4B-scale generative stack for efficient text-to-image generation and instruction-based image editing. The stack is built from two co-designed components: Mage-VAE, a lightweight high-fidelity latent tokenizer, and a Native-Resolution Multimodal Diffusion Transformer trained with rectified flow matching. Mage-VAE uses one-step diffusion-style encoding and decoding w
--
CS 6120: Loop Optimization
www.cs.cornell.edu
Also from Lesson 5, recall the loop-invariant code motion (LICM) optimization, which moves code from inside a loop to before the loop, if the computation always does the same thing on every iteration of the loop.
--
GitHub - rivet-dev/agentos: A faster, lighter, cheaper alternative to sandboxes. Run any coding agent inside an isolated Linux VM, with agent orchestration built in.
GitHub
A faster, lighter, cheaper alternative to sandboxes. Run any coding agent inside an isolated Linux VM, with agent orchestration built in. - rivet-dev/agentos
--