← Library

1. A model built for deterministic tasks - Interfaze

   Interfaze

   A new model architecture for deterministic tasks achieving the highest accuracy, precision and consistency for tasks like OCR, Audio understanding, Structured Data Extraction and more

2. AI 2040: Plan A

   AI 2040

   A detailed forecast and recommendation for how the US, China and the rest of the world should navigate superintelligence.

3. Neural Cellular Automata: From Cells to Pixels

   Ehsan Pajouheshgar1, · cells2pixels.github.io

   Explore interactive demos of a self-organizing pattern formation model that combines Neural Cellular Automata with Neural Fields.

4. GitHub - Emanuele-web04/Peekaboo: A tiny native macOS to-do list that reveals from your chosen screen corner.

   GitHub

   A tiny native macOS to-do list that reveals from your chosen screen corner. - Emanuele-web04/Peekaboo

5. GitHub - jxnl/personal-monorepo-template

   GitHub

   Contribute to jxnl/personal-monorepo-template development by creating an account on GitHub.

6. Introducing Plan A

   Scott Alexander · Astral Codex Ten

   ...

7. GitHub - stas00/ml-engineering: Machine Learning Engineering Open Book

   GitHub

   Machine Learning Engineering Open Book. Contribute to stas00/ml-engineering development by creating an account on GitHub.

8. SelRoute: Query-Type-Aware Routing for Long-Term Conversational Memory Retrieval

   Matthew McKee Independent Researcher Adelaide, Australia sindecker@gmail.com · arxiv.org

   Retrieving relevant past interactions from long-term conversational memory typically relies on large dense retrieval models (110M–1.5B parameters) or LLM-augmented indexing. We introduce SelRoute, a framework that routes each query to a specialized retrieval pipeline — lexical, semantic, hybrid, or vocabulary-enriched — based on its query type. On LongMemEval_M (Wu et al., 2024), SelRoute achieves Recall@5 of 0.800 with bge-base-en-v1.5 (109M parameters) and 0.786 with bge-small-en-v1.5 (33M par