How test-time scaling unlocks hidden reasoning abilities in small language models (and allows them to outperform LLMs)
A 1B small language model can beat a 405B large language model in reasoning tasks if provided with the right test-time scaling strategy.
Microsoft’s new AI agent can control software and robots
Microsoft Research introduced Magma, an integrated AI foundation model that combines visual and language processing to ...
IEEE5d
Fully Automated Software Product Line Evolution With Diverse Artifacts
Abstract: Existing approaches in software product lines usually neglect knowledge modeling and simulation of the interaction between features capable of bringing dynamism and automation. Consequently, ...