Skip to content
{ ken ashe }
  • Building
  • Topics
  • Blog
  • Building
  • Topics
  • Blog
← Blog / Tags

// tag

research

26 posts tagged research.

  • What Six Years of TrustNLP Papers Say About Where AI Safety Research Actually Went Aug 12, 2026
  • Agnostic PAC learning gets its optimal bound Aug 7, 2026
  • Sparse Weight Decomposition makes circuit extraction less expensive Aug 5, 2026
  • Teaching Models Formal Logic Before Words: What Logic-PPT Actually Shows Aug 5, 2026
  • GradCuit optimizes the reasoning state, not the model Aug 4, 2026
  • LiveMem reframes long-context memory as state continuity Aug 4, 2026
  • On-policy imitation helps when the student is smaller than the expert Aug 3, 2026
  • ReToken Makes Visual Retrieval a One-Token Routing Problem Jul 31, 2026
  • The Price of Monoculture: What Happens to Writing When Everyone Uses the Same Model Jul 30, 2026
  • MODUS brings any-to-any multimodal modeling to decoder-only systems Jul 29, 2026
  • πR² makes robot policies react inside the action chunk Jul 29, 2026
  • Hyperball optimizers still need learning-rate discipline Jul 27, 2026
  • Quantum Spectral Models: Encoding a Matrix's Structure Into the Circuit Itself Jul 27, 2026
  • Barzilai-Borwein’s superlinear convergence problem Jul 24, 2026
  • Maskability Index makes prompt choice less vibes-based Jul 23, 2026
  • Sarcasm detection needs the mismatch, not just the meme Jul 20, 2026
  • RoboTTT treats robot memory as trainable state, not a longer prompt Jul 17, 2026
  • Transformer circuits may be lower-dimensional than they look Jul 14, 2026
  • Transformer Theory Moves From Can Represent to Can Learn Jul 14, 2026
  • PHINN-EEG’s dream detection claim is a proposal, not a result Jul 13, 2026
  • Co-LMLM Puts Facts in a Database, Not the Weights Jul 9, 2026
  • ELSA3D routes language to the right 3D scale Jul 8, 2026
  • Graph attention needs the graph’s spectrum, not an average filter Jul 8, 2026
  • C2R targets the hidden mess inside sparse autoencoder features Jun 30, 2026
  • DiScoFormer treats density and score estimation as one reusable job Jun 30, 2026
  • Tapered Language Models and the Free Lever in Layer Width Jun 23, 2026
  • RSS
  • LinkedIn
  • X
  • GitHub
  • Email
  • Newsroom
  • Media Kit
  • Privacy
  • Terms
  • Disclosure

© 2026 Ken Ashe