Topics /
A NeurIPS 2026 paper with code claims that deleting a tiny, learned set of weights matches full fine-tuning, and forgets less. The code release is one commit old. Read more >>
BLA hardens apps so AI reverse-engineering agents burn their token budget on decoys. LASIK is a static tool that proves the decoys dead and removes them. Read more >>
A compiler turns Qwen3.5-4B's own layer types into a working RV32 processor, then runs Vercel's fx agent on it one token at a time. It is slow, and it opens pull requests. Read more >>
A benchmark that re-reads the patches other benchmarks already published and counts syntax units added and removed. Its own data shows why that is tricky. Read more >>
A 1,600-line patch on Meta's muse-gadget-sdk that points speech-to-text, chat, and text-to-speech at any three OpenAI-compatible servers. Read more >>
An alpha-stage runner for ternary LLMs: a Go supervisor around a static C++ engine, with pull, run, and serve commands and an OpenAI-compatible endpoint. Read more >>
bigarrow is a small Swift CLI and agent skill that points at windows and buttons with arrows, boxes, and signs. It never clicks, types, or captures. Read more >>
A hold-to-talk macOS assistant that shells out to claude -p for hard requests, then saves the steps so the same request skips the model next time. Read more >>
A browser extension that summarizes pages, videos, and PDFs with small models on WebGPU or WebAssembly, with no account, API key, or backend. Read more >>