Records found: 4
'Skala is a neural exchange–correlation functional from Microsoft Research that achieves hybrid-level accuracy at semi-local cost and is available on GitHub and via Azure AI Foundry.'
Microsoft and Tsinghua researchers propose Reward Reasoning Models that adaptively allocate compute resources during evaluation, significantly improving large language model judgment and alignment across complex tasks.
New research from Microsoft and Salesforce shows that large language models experience a 39% performance drop when handling real multi-turn conversations with incomplete instructions, highlighting a key challenge in conversational AI.
'Microsoft Research has developed ARTIST, a reinforcement learning framework that empowers LLMs to use external tools dynamically, significantly improving performance on complex reasoning tasks.'