MARBLE: Multi-Agent Reasoning for Bioinformatics Learning and Evolution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Sunghyun, Yun, Seokwoo, Yun, Youngseo, Lee, Youngrak, Lim, Sangsoo
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918298178813952
author Kim, Sunghyun
Yun, Seokwoo
Yun, Youngseo
Lee, Youngrak
Lim, Sangsoo
author_facet Kim, Sunghyun
Yun, Seokwoo
Yun, Youngseo
Lee, Youngrak
Lim, Sangsoo
contents Motivation: Developing high-performing bioinformatics models typically requires repeated cycles of hypothesis formulation, architectural redesign, and empirical validation, making progress slow, labor-intensive, and difficult to reproduce. Although recent LLM-based assistants can automate isolated steps, they lack performance-grounded reasoning and stability-aware mechanisms required for reliable, iterative model improvement in bioinformatics workflows. Results: We introduce MARBLE, an execution-stable autonomous model refinement framework for bioinformatics models. MARBLE couples literature-aware reference selection with structured, debate-driven architectural reasoning among role-specialized agents, followed by autonomous execution, evaluation, and memory updates explicitly grounded in empirical performance. Across spatial transcriptomics domain segmentation, drug-target interaction prediction, and drug response prediction, MARBLE consistently achieves sustained performance improvements over strong baselines across multiple refinement cycles, while maintaining high execution robustness and low regression rates. Framework-level analyses demonstrate that structured debate, balanced evidence selection, and performance-grounded memory are critical for stable, repeatable model evolution, rather than single-run or brittle gains. Availability: Source code, data and Supplementary Information are available at https://github.com/PRISM-DGU/MARBLE.
format Preprint
id arxiv_https___arxiv_org_abs_2601_14349
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MARBLE: Multi-Agent Reasoning for Bioinformatics Learning and Evolution
Kim, Sunghyun
Yun, Seokwoo
Yun, Youngseo
Lee, Youngrak
Lim, Sangsoo
Multiagent Systems
Machine Learning
Motivation: Developing high-performing bioinformatics models typically requires repeated cycles of hypothesis formulation, architectural redesign, and empirical validation, making progress slow, labor-intensive, and difficult to reproduce. Although recent LLM-based assistants can automate isolated steps, they lack performance-grounded reasoning and stability-aware mechanisms required for reliable, iterative model improvement in bioinformatics workflows. Results: We introduce MARBLE, an execution-stable autonomous model refinement framework for bioinformatics models. MARBLE couples literature-aware reference selection with structured, debate-driven architectural reasoning among role-specialized agents, followed by autonomous execution, evaluation, and memory updates explicitly grounded in empirical performance. Across spatial transcriptomics domain segmentation, drug-target interaction prediction, and drug response prediction, MARBLE consistently achieves sustained performance improvements over strong baselines across multiple refinement cycles, while maintaining high execution robustness and low regression rates. Framework-level analyses demonstrate that structured debate, balanced evidence selection, and performance-grounded memory are critical for stable, repeatable model evolution, rather than single-run or brittle gains. Availability: Source code, data and Supplementary Information are available at https://github.com/PRISM-DGU/MARBLE.
title MARBLE: Multi-Agent Reasoning for Bioinformatics Learning and Evolution
topic Multiagent Systems
Machine Learning
url https://arxiv.org/abs/2601.14349