arXiv cs.CL· Gabriel Asher, Cailum M. K. Stienstra, Tomo Oga, Devesh Shah Antonio H. de O. Fonseca, Niall O'Connor, Michael Widrich·· 11 小时前AI 评分38
LLM 在 MS/MS 结构解析的分布外泛化中依赖专用工具
LLMs Depend on Specialist Tools for Out-of-Distribution Generalization in MS/MS Structural Elucidation
AI 导读
前沿 LLM 在 MS/MS 从头结构解析中分布内 Top-1 准确率可达 52%，但分布外仅 2%。研究基于 MassSpecGym 构建 100 种分析物基准，仅接入 de novo 结构生成器可将分布外准确率提升至最高 38%，且所得智能体在分布内外均优于独立 de novo 模型。
正文 · 原文
Abstract:A frontier language model was recently reported to solve 1D NMR structural elucidation without tools. We ask whether this capability transfers to tandem mass spectrometry (MS/MS) in the de novo setting, where the model must propose a structure directly from the spectrum without a candidate list or reference library. To separate reasoning from recall, we build a 100-analyte benchmark from MassSpecGym, split by each analyte's textual footprint into in-distribution (well documented in literature) and out-of-distribution (essentially undocumented) arms. We hold the model, harness, and output format fixed and vary only the LLM's access to domain-specific tools. Without tools, a frontier LLM reaches 52% Top-1 accuracy in-distribution but only 2% out-of-distribution. Only access to de novo structure generators improves out-of-distribution accuracy, raising it to as much as 38% Top-1, and the resulting agent outperforms the standalone de novo models both in and out of distribution. These results suggest that LLMs are not effective at de novo MS/MS structural elucidation on their own, and that further gains depend on the specialist tools available to the agent rather than on the LLM itself.
| Comments: | Accepted at "NeurIPS 2026 Workshop on Agentic Systems for Molecular Sciences" and "I Can't Believe It's Not Better: Failure Modes of AI in Biology Workshop at NeurIPS 2026" |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (stat.ML) |
| MSC classes: | 68T07, 68T50, 68T42, 92E10 |
| ACM classes: | I.2.6; I.2.7; I.2.1; I.2.11; J.2; J.3 |
| Cite as: | arXiv:2611.03308 [cs.LG] |
| (or arXiv:2611.03308v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.03308 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Michael Widrich [view email]
[v1]
Mon, 2 Nov 2026 15:39:07 UTC (311 KB)
来源：arXiv cs.CL · arxiv.org