Translation

Bigger models are not always better, as A*STAR researchers show how a targeted AI model can beat global, general-purpose models in summarising multilingual regional news.

Obviating the need for large datasets in machine translation models could be a reality, thanks to a new method for diversifying training data

A*STAR researchers have devised a machine learning strategy that can be applied to translation and other complex classification problems.

DNA previously
considered superfluous may provide vital protection against stress and could
even be used in cancer treatment

Computer simulations help design molecules with potential as a cancer therapy

Researchers identify a genetic switch that controls differentiation of embryonic stem cells into neural cells

Zebrafish study reveals a gene regulatory network responsible for establishing and maintaining muscle fiber lineage

The crystal structure of the Dom34–Hbs1 complex reveals how cells recognize and remove defective messenger RNA with stalled elongation through ‘no-go’ decay