Gemini多智能体会被同伴错误带偏（38%案例），独立裁判能抓，门控不行，做临床AI的注意了。
一项研究测试了临床多智能体委员会是否会被捷径欺骗，覆盖文本、影像和ICU表格七组队列。在MedQA-USMLE等六个数据集上，Gemini委员会单独面对误导线索时仅有5-16%的翻转率，但两个同伴给出同一错误答案时，holdout模型在38%案例中采纳。三种监督智能体中，gate的假阳性率达100%，同源judge在文本上精确率100%但在影像上失效，而独立referee在影像上取得77-88%精确率。研究还发现，视觉显著性提高三倍不增加传染，但第二个同伴声音使传染率提高一半。
Agents Catching Agents: Shortcut Cascades and Benchmark Gaming in Clinical Multi-Agent Systems
Clinical decision support is moving toward committees of language-model agents deliberating on a shared workspace. We ask whether such committees can be gamed by shortcuts, cues a benchmark rewards but a clinician would ignore. Across seven cohorts on six public datasets spanning text (MedQA-USMLE, MedMCQA, MIMIC-CXR reports), imaging (NIH ChestX-ray14, MIMIC-CXR-JPG, CheXpert) and tabular ICU records (SUPPORT2), Gemini committees resist these cues in isolation (flip 5-16%), yet a socially plausible shortcut spreads: when two peers assert the same wrong answer, the holdout under test adopts it in 38% of cases, as does a false "pre-screen" system flag, on both capability tiers. Of three oversight agents, a gate cannot separate adoption from honest agreement (false-positive rate 100%); a same-lineage judge reading only the transcript flags adoption on text (precision 100%, recall 93%) but collapses onto the gate in imaging; a referee that privately re-queries the holdout transfers to imaging (77-88% precision, 13-21% false-positive rate). Tripling a cue's visual salience does not move contagion, whereas a second peer voice raises it by half again. Gaming a hidden rubric is near-silent: only 1/10 text and 1/134 imaging drifters name the rubric they moved toward. What games a committee is social plausibility, and only a referee independent of self-report catches it. Code: https://github.com/criticaldata/benchmaxxing