Track new results across pure and applied mathematics.
Search papers, labs, and topics across Lattice.
This paper addresses the computational inefficiency of de novo nanobody design by recasting condition selection (scaffold and epitope) as an inference-time resource allocation problem, moving beyond single-configuration steering. The authors introduce two dynamic budget allocators, Global Sequential Monte Carlo (SMC) and Batched Successive Elimination (BaSE), to concurrently steer generation and allocate compute across multiple design configurations. These methods achieve a nearly 5x increase in designable nanobody candidates per GPU-hour without compromising diversity.
Nanobody design yield jumps 5x by dynamically allocating compute across scaffolds and epitopes at inference time, eliminating the need for upfront configuration commitment.
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.