Primary sourcearXiv cs.LG
View sourceEvent ·Single source
TxBench-PP基准测试：AI Agent在临床前药理学任务中表现不佳
TxBench-PP是一个用于评估AI agent在小分子临床前药理学中决策能力的基准，包含100个涉及作用机制、药效学等任务的评估。在16个模型配置（涉及11个模型和4800条轨迹）中，最佳配置Claude Opus 4.8 / Pi仅通过59.3%（178/300）的端点尝试，GPT-5.5 / Pi通过55.3%。结果表明，当前AI系统无法可靠复现临床前药理学决策。
Current read
想看看AI在药物发现中到底行不行？这个基准测试用4800条轨迹告诉你，Claude Opus 4.8和GPT-5.5都还差得远，最高才59.3%的通过率。
2 sources47° AI heatLast updated 2026/6/17 16:23:45
Evidence
2 sourcesRelated source 1@hebbia
View sourceConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.