研究评估智能体设计治疗性抗体的可靠性
Can agents reliably design therapeutic antibodies?
这篇论文测了智能体在 10 个专家验证的抗体亲和力任务上到底靠不靠谱，还对比了工具和历史记录两种增强的取舍。
一项 arXiv 研究考察智能体在抗体设计中的表现，区分 task-guided（给定分子编辑指令）和 black-box（仅依据已评分序列提出改进）两种设定。研究者整理了 10 个经专家验证的亲和力改进任务，发现 LLM 评判器能近似复现程序化的指令遵循评估。实验显示：给智能体接入抗体相关工具可提升指令遵循和序列空间探索，但降低优化性能；引入设计历史则相反。将两种设定结合到 in silico 优化循环后，智能体优化器与成熟的贝叶斯优化基线表现相当。
Can agents reliably design therapeutic antibodies?
Recent work has demonstrated agents driving protein design workflows, either through orchestration of specialized tools or direct manipulation of sequences. Yet the fidelity with which agents follow protein design instructions and the factors contributing to their effectiveness remain understudied. We examine agentic antibody design under two settings: when explicit molecular-editing instructions are provided and when agents receive only previously-scored sequences and are prompted to propose improvements. We call the first task-guided and the second black-box design. For the task-guided setting, we curate ten expert-validated affinity-improvement tasks and show that LLM judges closely reproduce programmatic evaluations of instruction following. We then study two augmentations: access to antibody-relevant tools and to previously scored design history. Tools improve instruction following (IF) and sequence space exploration while reducing optimization performance, whereas design history reduces IF but improves optimization performance at the expense of exploration. Finally, we combine taskguided and black-box design within an in silico optimization loop and show that our agentic optimizer performs comparably to an established Bayesian optimization baseline.
