IRIS: Interactive Research Ideation System for Accelerating Scientific Discovery

2025-04-23

AI Summary

To address the low efficiency, poor transparency, and weak controllability in hypothesis generation during early-stage scientific research, this paper proposes a human-in-the-loop (HITL)-driven interactive scientific ideation system. Methodologically, it introduces the first adaptive test-time computation expansion mechanism integrated with Monte Carlo Tree Search (MCTS), coupled with a fine-grained user feedback loop and query-driven semantic synthesis of scholarly literature—enabling highly transparent and tightly controllable hypothesis generation. Experimental results demonstrate that the system significantly improves the novelty, feasibility, and insight depth of hypotheses formulated by interdisciplinary researchers. The open-source implementation has been widely adopted by the research community.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Humans and AI: Human-Computer Interaction

Application Category

• Economics, Online Markets and Human Computation: Research challenges in human and human-AI computation
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The rapid advancement in capabilities of large language models (LLMs) raises a pivotal question: How can LLMs accelerate scientific discovery? This work tackles the crucial first stage of research, generating novel hypotheses. While recent work on automated hypothesis generation focuses on multi-agent frameworks and extending test-time compute, none of the approaches effectively incorporate transparency and steerability through a synergistic Human-in-the-loop (HITL) approach. To address this gap, we introduce IRIS: Interactive Research Ideation System, an open-source platform designed for researchers to leverage LLM-assisted scientific ideation. IRIS incorporates innovative features to enhance ideation, including adaptive test-time compute expansion via Monte Carlo Tree Search (MCTS), fine-grained feedback mechanism, and query-based literature synthesis. Designed to empower researchers with greater control and insight throughout the ideation process. We additionally conduct a user study with researchers across diverse disciplines, validating the effectiveness of our system in enhancing ideation. We open-source our code at https://github.com/Anikethh/IRIS-Interactive-Research-Ideation-System

Problem

• How can LLMs accelerate scientific discovery effectively?
• Generating novel hypotheses with human-in-the-loop transparency.
• Enhancing ideation via adaptive compute and feedback mechanisms.

Innovation

• Monte Carlo Tree Search for adaptive compute expansion
• Fine-grained feedback mechanism for ideation
• Query-based literature synthesis for research