TrustResearcher: Automating Knowledge-Grounded and Transparent Research Ideation with Multi-Agent Collaboration

TL;DR

TrustResearcher introduces a four-stage, multi-agent framework for knowledge-grounded and transparent research ideation, bridging LLM-enabled brainstorming with traceability and expert oversight. It integrates Structured Knowledge Curation, Diversified Idea Generation, Multi-stage Idea Selection, and Expert Panel Review & Synthesis, coupling grounding, iterative self-refinement, and optional human-in-the-loop. Demonstrated on a $k$-truss breaking case study, the system generates diverse, evidence-aligned hypotheses with intermediate reasoning traces and experimental critiques, plus a live demo and open-source code. This work provides a domain-agnostic blueprint for trustworthy AI-assisted scientific ideation and reproducible knowledge-driven proposal generation.

Abstract

Effective research relies on organizing extensive information and stimulating novel solutions. Agentic systems have recently emerged as a promising tool to automate literature-based ideation. However, current systems often remain black-box. Their outputs may appear plausible but weakly grounded, with limited transparency or control for researchers. Our work introduces TrustResearcher, a multi-agent demo system for knowledge-grounded and transparent ideation. Specifically, TrustResearcher integrates meticulously designed four stages into a unified framework: (A) Structured Knowledge Curation, (B) Diversified Idea Generation, (C) Multi-stage Idea Selection, and (D) Expert Panel Review & Synthesis. Different from prior pipelines, our system not only exposes intermediate reasoning states, execution logs, and tunable agents for inspections, but also enables the generation of hypotheses that are both diverse and evidence-aligned. Our design is also domain-agnostic: as long as literature sources exist, the same pipeline can be instantiated in any scientific field. As an illustrative case, we demonstrate TrustResearcher on a graph-mining case study (k-truss breaking problem), where it generates distinct, plausible hypotheses with evidence and critiques. A live demo and source code are available at https://github.com/valleysprings/TrustResearcher.

Figures (2)

• Figure 1: System architecture of TrustResearcher, illustrated with the $k$-truss breaking problem. The system comprises four modules connected in an end-to-end pipeline. Solid/Hollow arrows denote steps handled by agents.
• Figure 2: System interface during a live demonstration on the $k$-truss breaking problem.