Reproduce “Position Paper: Towards Open Complex Human-AI Agents Collaboration Systems for Problem Solving and Knowledge Management” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ju Wu, Calvin K.L. Or
A comprehensive theoretical framework and architectural blueprint, the Hierarchical Exploration–Exploitation Net (HE²-Net), is introduced for Human–AI Agents Collaboration Systems. It leverages Interpreted Petri Nets to integrate human and AI strengths for complex problem-solving and knowledge management, establishing a principled approach to agent initiative, knowledge validation, and system governance.
Key insights
- Agent identity and reconfigurability in HAACS can be rigorously defined through a boundary-centric ontology that synthesizes classic cybernetics and interface-oriented views.
- Formal modeling with hierarchical Interpreted Petri Nets enables explicit representation and control of concurrency, ownership, and complex interactions in human-AI systems, moving beyond ad-hoc designs.
- Effective collaboration in open-ended problems requires a meta-level "explore-exploit" principle to dynamically balance resource allocation between leveraging validated knowledge and generating new, provisional knowledge.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.