谷歌用模拟住院医训练Gemini，诊断准确率升至88%
Thats the reaserach i love to see: Google put Gemini through a simulated medical…
Thats the reaserach i love to see: Google put Gemini through a simulated medical residency, and it became markedly better at conducting clinical consultations.
这正是我喜欢看到的研究：谷歌让Gemini经历了一次模拟医学住院医师培训，它在进行临床咨询方面的能力显著提升。
ResidencyRL trained Gemini 3.5 Flash across 49,870 simulated telehealth encounters covering 81 conditions.
ResidencyRL在涵盖81种病症的49,870次模拟远程医疗接诊中训练了Gemini 3.5 Flash。
The AI patients hid symptoms, resisted advice and requested inappropriate treatments, forcing the model to gather information over conversations of up to 60 turns.
AI患者会隐藏症状、抗拒建议并要求不适当的治疗，迫使模型在长达60轮的对话中收集信息。
After training, diagnostic accuracy under adversarial conditions rose from 81% to 88%, while missed red flags fell by 31%. In a blinded evaluation of 97 cases, clinicians preferred the trained agent over the base model in 87.6% of comparisons.
训练后，在对抗条件下的诊断准确率从81%提升到88%，漏诊危险信号的情况下降了31%。在97个病例的盲评中，临床医生在87.6%的比较中更偏好训练后的智能体而非基础模型。
The improvement also transferred to unseen oncology cases and external benchmarks.
这种改进也迁移到了未见过的肿瘤病例和外部基准测试中。
AI health agents incoming <3
AI健康智能体即将到来 <3
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力