Skip to content
{ ken ashe }
  • Building
  • Writing
  • About
  • Newsroom
  • Digest
  • Building
  • Writing
  • About
  • Newsroom
  • Digest
← Digest / Tags

// tag

language-models

23 posts tagged language-models.

  • Models may refuse based on who they think you are Sep 28, 2026
  • What LLMs Miss About Haitian Creole, and Why Low-Resource Culture Breaks Evals Sep 28, 2026
  • Audio models do not automatically share speech and text concepts Sep 25, 2026
  • MISVO steers frozen language models at inference time Sep 25, 2026
  • When a model’s rejection reason changes its next choice Sep 25, 2026
  • Authorship verification works better as similarity, not classification Sep 24, 2026
  • Translation Fine-Tuning Breaks the Controls General Evals Miss Sep 24, 2026
  • Speaker-Centered Memory: Why Group Chats Break Your AI Agent Sep 23, 2026
  • Membership Inference Gets an Entropy Correction: Reading the ETD Paper Sep 21, 2026
  • Softmax attention needs an off switch Sep 21, 2026
  • ComPO and the Case Against Optimizing the Loss You Wrote Down Sep 17, 2026
  • What Actually Makes a Tokeniser Good: Search Beats Objective Sep 17, 2026
  • Teaching Models to Say 'I Don't Know' With a Prompt, Not a Retrain Sep 16, 2026
  • A thin local AI signal still tells builders what to watch Sep 13, 2026
  • When a Language Model's Odds Don't Add Up Sep 3, 2026
  • ConvergeFlow's Bet: Kill the Decoder, Trust the Flow Aug 25, 2026
  • Mimir v1 tests whether small open models can win on clean data Aug 14, 2026
  • Entity matching needs architecture tests, not bigger-model reflexes Jul 28, 2026
  • Maskability Index makes prompt choice less vibes-based Jul 23, 2026
  • Language model embeddings do not want to collapse Jul 13, 2026
  • Co-LMLM Puts Facts in a Database, Not the Weights Jul 9, 2026
  • DiaLLM separates dialect understanding from dialect writing Jul 9, 2026
  • Safety evals need to test messy language, not just bad behavior Jul 2, 2026

Ken Ashe ·AI application builder ·CPA ·PMP

Building with AI in public. No hype, no doom. Receipts only.

hello@kenashe.ai

Explore

  • Building
  • Writing
  • Digest
  • Topics

About & Press

  • About
  • Newsroom
  • Media Kit
  • Lucky Domains

Social

  • LinkedIn
  • X
  • GitHub
  • RSS

Legal

  • Privacy
  • Terms
  • Disclosure

© 2026 Ken Ashe ·Built with AI in public