核物理研究者终于有了一个统一的多任务预测工具——用 LLM 微调替代传统多模型方案，精度提升显著且效率更高，做核数据分析和理论验证的团队值得关注。
研究人员提出了一种基于大语言模型的多任务学习框架，用于统一描述多个核物理可观测量。通过低秩适配（LoRA）微调预训练的 DeepSeek-R1-1.5B 模型，在保留通用参数的同时引入轻量适配器。该模型在因果语言建模范式下，对实验值与理论值的偏差进行自回归训练，在电荷半径、质量、结合能、分离能和衰变能等七个可观测量的预测上实现了显著精度提升，训练损失在所有任务中下降超过 98%。这项工作展示了基于 LLM 的框架通过结构化先验嵌入，为核物理基础属性的多任务回归提供了一种高效且共享的方法。
Large language model for unified and accurate description of multidimensional nuclear properties
A prior-informed large language model (LLM) driven multi-task learning framework is proposed for the unified description of multiple nuclear observables. By fine-tuning the pre-trained DeepSeek-R1-1.5B model with Low-Rank Adaptation (LoRA), lightweight adapters are introduced while preserving general pre-trained parameters. Under a causal language modeling paradigm, the model is trained autoregressively on deviations between experimental and theoretical values. Significant accuracy improvements are achieved across seven observables, including charge radii, masses, binding energies, separation energies, and decay energies, with the training loss decreasing by over 98% across all tasks. This demonstrates that the LLM-based framework, through structured prior embedding, offers an efficient and shared approach for multi-task regression in fundamental nuclear properties.