# ACED: Revolutionizing Instrumental Analysis Using Foundation Models

> **NSF 01002526DB NSF RESEARCH & RELATED ACTIVIT** · University of Massachusetts Amherst (MA) · $500,000

## Abstract

Decoding the structures and properties of unknown molecules through analyzing the wavelengths of their electromagnetic properties is known as spectral analysis. Spectral analysis is crucial for scientific discovery and practical applications in various fields, including material manufacture, drug design, food safety, explosive detection, and non-invasive diagnosis. Spectral analysis offers rapid, sensitive, non-destructive, and cost-effective identification of unknown molecules through their characteristic numerical signals and outperforms traditional chemical analysis. However, the process of translating numerical signals into molecular structures is currently resource-intensive and not user-friendly because it often requires extensive trial-and-error and specialized training. This project aims to revolutionize spectral analysis using state-of-the-art artificial intelligence (AI) in an automatic, accelerated, and accurate fashion. This project will treat spectral signals and molecular structures as two different "languages". Models developed in this project will automatically transform spectral signals and molecular structures into descriptions of molecules in the two languages and enable rapid conversion between each description based on advanced AI-powered language translation tools. The resulting universal toolkit will simplify and streamline spectral analysis in practical scenarios and benefit applications in scientific research, national healthcare, national security, e

## Key facts

- **NSF award ID:** 2435822
- **Awardee organization:** University of Massachusetts Amherst (MA)
- **SAM.gov UEI:** VGJHK59NMPK9
- **PI:** Zhou Lin
- **Primary program:** 01002526DB NSF RESEARCH & RELATED ACTIVIT
- **All programs:** —
- **Estimated total:** $500,000
- **Funds obligated:** $500,000
- **Transaction type:** Standard Grant
- **Period:** 07/01/2025 → 06/30/2027

## Primary source

NSF Award Search: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2435822

## Citation

> US National Science Foundation, Award 2435822, ACED: Revolutionizing Instrumental Analysis Using Foundation Models. Retrieved via AI Analytics 2026-06-07 from https://api.ai-analytics.org/grant/nsf/2435822. Licensed CC0.

---

*[NSF Awards dataset](/datasets/nsf-awards) · CC0 1.0*