大语言模型在生物分子建模中的局限性：跨尺度评估
The limits of bio-molecular modeling with large language models : a cross-scale evaluation
AI总结：
本文通过跨尺度生物分子基准测试，揭示了大语言模型在生物分子建模中的性能与机制理解之间的差距，指出链式思维数据对生物任务的有限帮助，混合mamba-注意力架构在长序列中的有效性，以及监督微调对专业化与泛化能力的影响。
摘要：
跨分子尺度的生物分子系统建模仍然是科学研究中的核心挑战。尽管大语言模型（LLMs）越来越多地应用于生物分子发现，但对多尺度生物问题的系统评估和对其工具增强能力的严格评估仍有限。通过提出的跨尺度生物分子基准测试：BioMol-LLM-Bench，一个包含26个下游任务、覆盖4个不同难度级别的统一框架，以及整合的计算工具，实现了更全面的评估。在13个代表性模型上的评估揭示了四个主要发现：链式思维数据提供有限的益处，甚至可能降低生物任务的性能；混合mamba-注意力架构在长生物分子序列中更有效；监督微调提高了专业化，但牺牲了泛化能力；当前LLMs在分类任务上表现良好，但在具有挑战性的回归任务上仍较弱。这些发现为未来基于LLM的分子系统建模提供了实用指导。
英文摘要：
The modeling of bio-molecular system across molecular scales remains a central challenge in scientific research. Large language models (LLMs) are increasingly applied to bio-molecular discovery, yet systematic evaluation across multi-scale biological problems and rigorous assessment of their tool-augmented capabilities remain limited. We reveal a systematic gap between LLM performance and mechanistic understanding through the proposed cross-scale bio-molecular benchmark: BioMol-LLM-Bench, a unified framework comprising 26 downstream tasks that covers 4 distinct difficulty levels, and computational tools are integrated for a more comprehensive evaluation. Evaluation on 13 representative models reveals 4 main findings: chain-of-thought data provides limited benefit and may even reduce performance on biological tasks; hybrid mamba-attention architectures are more effective for long bio-molecular sequences; supervised fine-tuning improves specialization at the cost of generalization; and current LLMs perform well on classification tasks but remain weak on challenging regression tasks. Together, these findings provide practical guidance for future LLM-based modeling of molecular systems.
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。
