Saved in:
Bibliographic Details
Main Authors: Pu, Yingming, Lin, Tao, Chen, Hongyu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.15047
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908821269512192
author Pu, Yingming
Lin, Tao
Chen, Hongyu
author_facet Pu, Yingming
Lin, Tao
Chen, Hongyu
contents Large Language Model (LLM)-based multi-agent systems (MAS) demonstrate remarkable potential for scientific discovery. Existing approaches, however, often automate scientific discovery using predefined workflows that lack rationality constraints. This often leads to aimless hypothesizing and a failure to consistently link hypotheses with evidence, thereby hindering the systematic reduction of uncertainty. Overcoming these limitations fundamentally requires a principled approach to exploration. We introduce PiFlow, an information-theoretical framework, treating automated scientific discovery as a structured uncertainty reduction problem guided by principles (e.g., scientific laws). Extensive evaluations across three distinct scientific domains demonstrate that PiFlow (I) improves discovery efficiency by 31.18%~41.73% and solution quality by 12.47%~31.72% against state-of-the-art methods, (II) delivers a 5.6x speedup in time-to-solution while reducing token consumption by up to 27% compared to vanilla agents, and (III) serves as a Plug-and-Play module that generalizes on existing agent architecture. Overall, PiFlow establishes a novel paradigm shift in highly efficient agentic scientific discovery, paving the way for more robust and accelerated AI-driven research.
format Preprint
id arxiv_https___arxiv_org_abs_2505_15047
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PiFlow: Principle-Aware Scientific Discovery with Multi-Agent Collaboration
Pu, Yingming
Lin, Tao
Chen, Hongyu
Machine Learning
Artificial Intelligence
Large Language Model (LLM)-based multi-agent systems (MAS) demonstrate remarkable potential for scientific discovery. Existing approaches, however, often automate scientific discovery using predefined workflows that lack rationality constraints. This often leads to aimless hypothesizing and a failure to consistently link hypotheses with evidence, thereby hindering the systematic reduction of uncertainty. Overcoming these limitations fundamentally requires a principled approach to exploration. We introduce PiFlow, an information-theoretical framework, treating automated scientific discovery as a structured uncertainty reduction problem guided by principles (e.g., scientific laws). Extensive evaluations across three distinct scientific domains demonstrate that PiFlow (I) improves discovery efficiency by 31.18%~41.73% and solution quality by 12.47%~31.72% against state-of-the-art methods, (II) delivers a 5.6x speedup in time-to-solution while reducing token consumption by up to 27% compared to vanilla agents, and (III) serves as a Plug-and-Play module that generalizes on existing agent architecture. Overall, PiFlow establishes a novel paradigm shift in highly efficient agentic scientific discovery, paving the way for more robust and accelerated AI-driven research.
title PiFlow: Principle-Aware Scientific Discovery with Multi-Agent Collaboration
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2505.15047