Towards a Unified Textual Graph Framework for Spectral Reasoning via Physical and Chemical Information Fusion

TL;DR

This paper introduces a multi-modal spectral analysis framework that combines physical spectral data with chemical knowledge using textual graphs and large language models, improving interpretability and generalization.

Contribution

It presents a novel unified textual graph framework that integrates spectral measurements with chemical semantics for enhanced spectral reasoning.

Findings

High performance across multiple spectral tasks

Robust generalization in zero-shot and few-shot scenarios

Effective multi-modal integration with minimal manual annotation

Abstract

Motivated by the limitations of current spectral analysis methods-such as reliance on single-modality data, limited generalizability, and poor interpretability-we propose a novel multi-modal spectral analysis framework that integrates prior knowledge graphs with Large Language Models. Our method explicitly bridges physical spectral measurements and chemical structural semantics by representing them in a unified Textual Graph format, enabling flexible, interpretable, and generalizable spectral understanding. Raw spectra are first transformed into TAGs, where nodes and edges are enriched with textual attributes describing both spectral properties and chemical context. These are then merged with relevant prior knowledge-including functional groups and molecular graphs-to form a Task Graph that incorporates "Prompt Nodes" supporting LLM-based contextual reasoning. A Graph Neural Network further processes this structure to complete downstream tasks. This unified design enables seamless multi-modal integration and automated feature decoding with minimal manual annotation. Our framework achieves consistently high performance across multiple spectral analysis tasks, including node-level, edge-level, and graph-level classification. It demonstrates robust generalization in both zero-shot and few-shot settings, highlighting its effectiveness in learning from limited data and supporting in-context reasoning. This work establishes a scalable and interpretable foundation for LLM-driven spectral analysis, unifying physical and chemical modalities for scientific applications.