RamanPFN：面向拉曼光谱 in-context 学习的物理引导光谱基础框架
RamanPFN: A physics-guided spectral foundation framework for in-context learning across diverse Raman applications
AI 导读
研究者提出通用光谱基础框架 RamanPFN，通过物理引导的谱学习实现统一 in-context 推理，编码光谱加性混合与振动带局部性，无需更新参数即可预测。在覆盖超 330,000 条光谱的 74 个公开拉曼数据集上，RamanPFN 在 129 个回归目标上优于化学计量学、光谱神经网络等 26 种方法，骨干 RMSE 降低 13.2%。
正文 · 原文
Abstract:Raman spectroscopy enables label-free molecular characterization across materials science, analytical chemistry, biomedicine, and industrial process monitoring. However, machine learning for high-dimensional spectroscopy remains constrained by limited labeled data and a mismatch between the physical organization of spectra and feature-agnostic models. Tabular foundation models such as TabPFN predict in context without parameter updates, but on wide spectra feature subsampling preserves channel coverage while losing joint visibility: in the five widest tasks, every channel was covered, yet only 4.4% of random channel pairs co-occurred in any forward pass. We present RamanPFN, a general-purpose spectral foundation framework for unified in-context inference through physics-guided spectral learning, which encodes the additive mixing of constituent spectra and the locality of vibrational bands. Global Compositional Unmixing decomposes distributed, multi-band mixture signatures into shared non-negative bases to couple covarying peaks across the full spectrum, while Local Vibrational Subspace Encoding extracts orthogonal modes within contiguous spectral neighborhoods to preserve peak morphology, intensity fluctuations, and peak shifts. Posterior marginalization over admissible unmixings unifies both scales into in-context prediction without updating the model. Across 74 public Raman datasets exceeding 330,000 spectra, RamanPFN outperformed 26 methods spanning chemometrics, spectral neural networks, deep tabular learners, and tabular foundation models on 129 regression targets, reducing backbone RMSE by 13.2%. We demonstrate across TabPFN-3, TabICLv2, and Kumo-Tabular that this advantage generalizes to both regression and classification tasks, reducing regression error by 12.3%-25.7% and multiclass log loss by 42.4%-63.5% to enable data-efficient quantitative analysis and sample identification.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2608.02157 [cs.LG] |
| (or arXiv:2608.02157v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2608.02157 arXiv-issued DOI via DataCite |
Submission history
From: Huan Wang [view email]
[v1]
Mon, 3 Aug 2026 12:37:55 UTC (934 KB)
[v2]
Wed, 30 Sep 2026 15:27:10 UTC (1,291 KB)
[v3]
Tue, 10 Nov 2026 13:34:47 UTC (1,485 KB)
来源：arXiv cs.AI · arxiv.org