研究内容の詳細は『BMJ』にて公開されました。 旧型AIを認知症診断で斬る 研究で実施された「モントリオール認知評価（MoCA）」テストにおける各AIモデルの得点を視覚的に示したグラフです。MoCAは最大30点満点で、通常、26点以上が正常範囲、26点未満は ...

Scientists tested publicly available LLM-driven chatbots including OpenAI's ChatGPT, Anthropic's Sonnet and Alphabet's Gemini using the Montreal Cognitive Assessment (MoCA) test — a series of tasks ...

Researchers arrived at this conclusion after evaluating the cognitive abilities of leading LLMs — ChatGPT versions 4 and 4o, Claude 3.5 ‘Sonnet’ (developed by Anthropic), and Gemini versions 1 and 1.5 ...

Methods: The VR cognitive assessment program and the traditional Montreal Cognitive Assessment (MOCA) test were performed on 12 patients with MCI and 108 healthy older adults. The VR program operates ...

In this paper, we propose MOCA (Multi-hypOtheses-based Curriculum learning for semi-supervised ASR), an accurate framework for semi-supervised ASR. MOCA generates multiple hypotheses for each speech ...

This repo serves as the official codebase for MoCA. We test MoCA on two backbone architectures, FastGAN and StyleGANv2. We refer to fast-gan-MoCA and StyleGANv2-MoCA for detailed instructions.