Rethinking Scientific Discovery in an Agentic Era

2026-07-04

AI Summary

This work proposes SCION, a scientific research operating system centered on scientific agents, addressing the fragmentation of existing AI4Science tools that struggle to autonomously orchestrate end-to-end scientific discovery. SCION translates high-level research intent into executable, auditable, and reusable workflows through Research Execution Plans (REPs). It introduces a novel goal-conditioned inverse search framework combined with batch active search strategies to enable organized, hierarchical, and memory-augmented multi-agent collaboration. Key technical innovations include a layered architecture, agent specialization, selective context construction, controlled delegation, and hierarchical cognitive memory. Evaluated on tasks spanning materials analysis, molecular design, and antibody screening, SCION demonstrates significant advantages over state-of-the-art methods in task decomposition, validation, optimization, and memory reuse.

Abstract

Artificial intelligence has advanced scientific discovery, but most AI4Science systems remain fragmented tools that rely on humans to coordinate problem formulation, literature grounding, model use, simulation, validation, and knowledge reuse. This paper presents \textbf{SCION (Scientific Collaborative Innovation with Agentic Organizational Nexus)}, an agentic scientific operating system that acts as an \textbf{organizational nexus}. Through a Science Agent serving as a \textbf{Meta-Harness}, SCION connects scientific tasks, tools, agents, artifacts, and memory, transforming research into an executable, auditable, and reusable operational process. At its core is the \textbf{Research Execution Plan (REP)}, which compiles high-level scientific intent into staged objectives, dependencies, verification checkpoints, tool requirements, expected artifacts, and fallback conditions. SCION further integrates hierarchical multi-agent execution, profile-driven specialization, selective context construction, governed delegation, and layered epistemic memory to support long-horizon scientific work. We formulate discovery under SCION as \textbf{Target-conditioned Inverse Search} and extend it to hidden-target settings through batch active search under finite experimental budgets. Applications in materials analysis, molecule design, and protein or antibody screening, together with experiments on scientific reading, idea generation, molecule generation, and antibody screening, show that SCION outperforms existing autonomous research-agent baselines, especially in decomposition, verification, refinement, and memory reuse. Overall, SCION shifts AI from isolated tools toward a coordinated operational layer for traceable and reusable scientific innovation.

Problem

• AI4Science
• scientific discovery
• agentic systems
• research automation
• knowledge reuse

Innovation

• Agentic Scientific Operating System
• Research Execution Plan (REP)
• Target-conditioned Inverse Search
• Hierarchical Multi-agent Coordination
• Epistemic Memory