[2509.12282] AISSISTANT: Human-AI Collaborative Review and Perspective Research Workflows in Data Science

Computer Science > Artificial Intelligence arXiv:2509.12282 (cs) [Submitted on 14 Sep 2025 (v1), last revised 1 Mar 2026 (this version, v2)] Title:AISSISTANT: Human-AI Collaborative Review and Perspective Research Workflows in Data Science Authors:Sasi Kiran Gaddipati, Farhana Keya, Gollam Rabby, Sören Auer View a PDF of the paper titled AISSISTANT: Human-AI Collaborative Review and Perspective Research Workflows in Data Science, by Sasi Kiran Gaddipati and 3 other authors View PDF HTML (experimental) Abstract:High-quality scientific review and perspective papers require substantial time and effort, limiting researchers' ability to synthesize emerging knowledge. While Large Language Models (LLMs) leverage AI Scientists for scientific workflows, existing frameworks focus primarily on autonomous workflows with very limited human intervention. We introduce AIssistant, the first open-source agentic framework for Human--AI collaborative generation of scientific perspectives and review research in data science. AIssistant employs specialized LLM-driven agents augmented with external scholarly tools and allows human intervention throughout the workflow. The framework consists of two main multi-agent systems: Research Workflow with seven agents and a Paper Writing Workflow with eight agents. We conducted a comprehensive evaluation with both human expert reviewers and LLM-based assessment following NeurIPS standards. Our experiments show that OpenAI o1 achieves the highest quality ...