DiMOS：基于 Doob 引导的免训练多目标科学设计搜索框架
DiMOS: Doob-Guided Inference-Time Multi-Objective Search for Scientific Design
科研朋友可以看看这篇：不用微调扩散模型，靠推理时搜索同时满足多个设计约束，6 个 DNA/蛋白/RNA 任务上联合成功率最高到基线 1.98 倍。
DiMOS 是一个免训练的推理时搜索框架，面向多目标科学设计任务，直接使用冻结的 masked diffusion 模型而无需微调。它通过候选补全的联合奖励执行近似 Doob 引导的局部重采样，绕开了 pass-or-fail 约束和黑盒奖励模型无法提供梯度的问题。为在有限推理预算内找到可行解，DiMOS 采用预算高效的轨迹搜索策略。在 DNA、蛋白质、RNA 共 6 个任务上，DiMOS 取得最高的联合成功率，在 DNA 和蛋白质任务上达到最强基线的 1.98 倍，同时保持较高的序列唯一性与自然度。
DiMOS: Doob-Guided Inference-Time Multi-Objective Search for Scientific Design
Scientific design often requires jointly satisfying multiple objectives and constraints. Pretrained masked diffusion models provide a generative foundation for this task, but fine-tuning them to meet these objectives and constraints incurs additional training costs, motivating inference-time guidance with frozen models. However, such guidance faces two challenges: pass-or-fail constraints and black-box reward models may provide no useful gradients, while jointly satisfying multiple requirements can leave a small feasible region, making feasible designs difficult to find within a limited inference budget. To address these challenges, we introduce DiMOS, a training-free framework for multi-objective scientific design. Using joint rewards from candidate completions, DiMOS performs approximate Doob-guided local resampling without requiring reward gradients. To allocate computation efficiently, it uses budget-efficient trajectory search to focus computation on promising continuations. Across six DNA, protein, and RNA tasks, DiMOS attains the highest joint success rate at comparable generation times, up to $1.98\times$ the strongest baseline on DNA and protein, while maintaining high sequence uniqueness and naturalness.
