大型发现模型：基于经验的基于模型的开放式搜索
Large Discovery Models: Empirically-grounded Model-Based Open-Ended Search
AI总结：
该研究提出大型发现模型（LDM），结合生成模型与贝叶斯非参数奖励替代模型，在神经网络训练、抗体设计和分子优化场景中，相比LLM反思或传统统计搜索，实现了更优的性能提升，可作为通用发现引擎。
AI中文摘要：
科学发现通常需要在庞大、结构化且开放式的假设空间（如分子、蛋白质序列、计算机程序）上优化评估成本高昂的目标。大型语言模型（LLMs）等生成模型为这类空间提供了表达性先验，但其似然性和自我评估并非目标的可靠替代，也无法提供校准后的认知不确定性，尤其是对于超出观测数据分布的新候选对象。我们提出大型发现模型（LDM），这是一种基于经验的循环架构，将生成模型与贝叶斯非参数奖励替代模型相结合。生成模型提出并优化候选设计，而替代模型预测其性能并量化不确定性，产生感知不确定性的价值以指导候选的生成、优化和选择。随着每次新的实验观测到来，发现记忆和替代模型会不断更新。我们在涵盖不同设计模态和目标的三个场景中评估LDM，包括神经网络训练、抗体设计和分子优化。与仅使用LLM反思或这些领域的传统统计搜索相比，LDM在验证BPB上实现了2.4倍的降低，结合能相对降低18.2%，且分子多目标性能获得超过60%的相对提升。这些结果表明，LDM可作为通用发现引擎，用于在开放式假设空间上进行有效搜索。
英文摘要：
Scientific discovery often involves optimising expensive-to-evaluate objectives over vast, structured, and open-ended hypothesis spaces, such as molecules, protein sequences, and computer programs. Generative models such as large language models (LLMs) provide expressive priors over such spaces, but their likelihoods and self-assessments are unreliable proxies for the objectives and calibrated epistemic uncertainty, especially for novel candidates outside the observed data distribution. We introduce the Large Discovery Model (LDM), an empirically grounded recurrent architecture that couples a generative model with a Bayesian non-parametric reward surrogate model. The generative model proposes and refines candidate designs, while the surrogate predicts their performance and quantifies uncertainty, yielding an uncertainty-aware value that guides candidate generation, refinement, and selection. The discovery memory and the surrogate model are continually updated as each new experimental observation arrives. We evaluate LDM on three scenarios spanning different design modalities and objectives, including neural-network training, antibody design, and molecular optimisation. Compared to LLM-only reflection or traditional statistical search across these domains, LDM achieves a $2.4\times$ greater reduction in validation BPB, an $18.2\%$ relative decrease in binding energy, and more than $60\%$ relative gains in molecular multi-objective performance. These results suggests that LDM could serve as a general-purpose discovery engine for effective search over open-ended hypothesis spaces.
