针对支架和表位的推理时引导的样本高效从头纳米抗体设计
Inference-Time Steering Across Scaffolds and Epitopes for Sample-Efficient De Novo Nanobody Design
AI总结：
该研究针对从头纳米抗体设计中条件选择导致的计算资源浪费问题，将生成式条件选择转化为推理时资源分配问题，提出两种动态预算分配器，可在不牺牲多样性的前提下大幅提升每GPU小时的可设计候选物产量。
摘要：
生成式分子设计通常依赖于过度生成与事后过滤。Feynman-Kac（FK）引导这类推理时引导方法通过将轨迹偏向高回报候选物来提升效率，但它们在单一固定配置内优化。在从头纳米抗体设计中，预先确定支架身份或目标表位等条件选择会造成瓶颈，由于不同条件下的产量存在差异，引导无前景的设置会浪费计算资源。我们将生成式条件选择重新表述为推理时资源分配问题，提出两种动态预算分配器：一种是联合全局序贯蒙特卡洛（SMC）公式，可同时在配置间进行引导与计算资源分配；另一种是解耦方法，采用适配的批量逐次消除（BaSE）算法。实验表明，这些动态分配器在不牺牲不同治疗靶点和生成配置多样性的前提下，每GPU小时可将可设计候选物的产量提升近5倍。
英文摘要：
Generative molecular design typically relies on over-generation and post-hoc filtering. Inference-time steering methods like Feynman-Kac (FK) steering improve efficiency by biasing trajectories toward high-reward candidates, yet they optimize within a single fixed configuration. In de novo nanobody design, upfront commitment to conditioning choices like scaffold identity or target epitope creates a bottleneck. Since yield varies across these conditions, steering unpromising setups wastes compute. We recast generative condition selection as an inference-time resource-allocation problem. We propose two dynamic budget allocators: a joint Global Sequential Monte Carlo (SMC) formulation that steers and allocates compute across configurations concurrently, and a decoupled approach adapting Batched Successive Elimination (BaSE). Empirically, these dynamic allocators increase the yield of designable candidates per GPU-hour nearly 5$\times$ without sacrificing diversity across distinct therapeutic targets and generative configurations.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
