PiFlow: Principle-Aware Scientific Discovery with Multi-Agent Collaboration

TL;DR

PiFlow introduces a principle-aware, information-theoretical framework for multi-agent scientific discovery that enhances efficiency, solution quality, and generalization by systematically reducing uncertainty guided by scientific principles.

Contribution

It presents PiFlow, a novel structured approach integrating scientific principles into multi-agent discovery, significantly improving efficiency and robustness over existing methods.

Findings

Improves discovery efficiency by 31.18% to 41.73%.

Reduces time-to-solution by 5.6x and token consumption by up to 27%.

Generalizes across existing agent architectures.

Abstract

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.