AI模型预测神经科学研究结果比人类专家,研究发现更准确。 AI models predict neuroscience study outcomes more accurately than human experts, study finds.

UCL的一项新研究表明,大型语言模型(LLMs)能够比人类专家更准确地预测神经科学研究结果。 A new study from UCL reveals that large language models (LLMs) can predict neuroscience study outcomes more accurately than human experts. 研究人员利用名为 " 脑 Bennch " 的工具发现,LLMs的精确率达到81%,而人类专家的准确率为63%。 Using a tool called BrainBench, researchers found that LLMs achieved an 81% accuracy rate, compared to 63% for human experts. 这表明人工智能可以通过帮助设计更有效的实验, 显著加速科学研究. This suggests that AI could significantly accelerate scientific research by helping design more effective experiments.