Guardado en:
Detalles Bibliográficos
Autores principales: Ramachandra, Nesar, Ting, Yuan-Sen, Sun, Zechang, Wells, Azton, Habib, Salman
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2508.10075
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915445994422272
author Ramachandra, Nesar
Ting, Yuan-Sen
Sun, Zechang
Wells, Azton
Habib, Salman
author_facet Ramachandra, Nesar
Ting, Yuan-Sen
Sun, Zechang
Wells, Azton
Habib, Salman
contents Pre-trained Large Language Models (LLMs) have revolutionized text processing, yet adapting Transformer-based neural networks to non-textual scientific modalities typically requires specialized architectures and extensive computational resources. We demonstrate that LLaMA-3.1-8B can be efficiently repurposed to predict galaxy redshifts from spectroscopic data through Low-Rank Adaptation (LoRA), achieving competitive performance while preserving its linguistic capabilities. Using only 16 GPU-hours and adapting 0.04% of model parameters, our approach achieves a mean absolute error of 0.04 in redshift prediction while retaining over 85% of performance on AstroBench and 89% on general QA tasks from eval-harness. This minimal-effort adaptation--requiring only simple standard fine-tuning APIs--lowers barriers to entry for domain scientists and enables integrated agentic workflows where a single model handles both spectroscopic data for quantitative analysis and natural language for reasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2508_10075
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Teaching LLMs to Speak Spectroscopy
Ramachandra, Nesar
Ting, Yuan-Sen
Sun, Zechang
Wells, Azton
Habib, Salman
Instrumentation and Methods for Astrophysics
Pre-trained Large Language Models (LLMs) have revolutionized text processing, yet adapting Transformer-based neural networks to non-textual scientific modalities typically requires specialized architectures and extensive computational resources. We demonstrate that LLaMA-3.1-8B can be efficiently repurposed to predict galaxy redshifts from spectroscopic data through Low-Rank Adaptation (LoRA), achieving competitive performance while preserving its linguistic capabilities. Using only 16 GPU-hours and adapting 0.04% of model parameters, our approach achieves a mean absolute error of 0.04 in redshift prediction while retaining over 85% of performance on AstroBench and 89% on general QA tasks from eval-harness. This minimal-effort adaptation--requiring only simple standard fine-tuning APIs--lowers barriers to entry for domain scientists and enables integrated agentic workflows where a single model handles both spectroscopic data for quantitative analysis and natural language for reasoning.
title Teaching LLMs to Speak Spectroscopy
topic Instrumentation and Methods for Astrophysics
url https://arxiv.org/abs/2508.10075