Search

7 results on this page

Claude Enters Protein Design

Claude Enters Protein Design

Claude is expanding into protein design, adding a scientific research capability to its existing AI portfolio. The update suggests potential use in computational biology and biotech workflows, though the article excerpt does not provide implementation or benchmark details.

The NeuronMedia9h ago#protein-design#scientific-ai
Transfer More Knowledge with Less Multilingual Data

Transfer More Knowledge with Less Multilingual Data

Apple Machine Learning presents a lexical-intervention approach for improving cross-lingual knowledge transfer when target-language data is scarce. The work targets downstream capabilities such as scientific reasoning, commonsense inference, and world knowledge without relying heavily on parallel data, translation systems, or auxiliary models.

Apple Machine LearningOfficial19h ago#multilingual-models
AI Can Solve More Math, But Not Invent It

AI Can Solve More Math, But Not Invent It

數學家袁新意表示,AI 近半年在整合既有數學思想、構造例子與反例方面進步迅速,但仍未產生足以讓職業數學家認可的重要原創成果。文章也指出,中國數學人才培養已出現質變,但青年教師承受過重的量化科研考核壓力。

🔬

Learning Machine Learning for Astronomy

A beginner asks for free resources, tutorials, books, and existing Jupyter notebooks for applying machine learning to astronomy. Potential use cases include analyzing JWST or TESS data to identify black-hole or exoplanet signatures and building reproducible environments with Git, Python libraries, and Docker.

Reddit r/MachineLearningCommunity1d ago#astronomy-ml#scientific-python#space-data
Page 1