INDIBATOR: Diverse and Fact-Grounded Individuality for Multi-Agent Debate in Molecular Discovery

TL;DR

INDIBATOR tackles the limits of coarse-grained personas in multi-agent molecular discovery by grounding each agent in a distinctive scientific DNA composed of publication history and molecular history. It introduces a three-phase debate (proposal, critique, voting) where individuality-guided agents iteratively generate and evaluate candidate molecules. Across three downstream tasks—protein-conditioned generation, bioactivity-guided design, and lead optimization—the approach yields higher performance and diversity than baselines relying on vanilla or keyword-based prompts, validating the value of fine-grained, fact-grounded individuality. Analyses reveal that granularity, diversity, and fact-grounding synergistically drive improvements, supporting broader applicability of individuality-grounded AI in domain-specific scientific discovery.

Abstract

Multi-agent systems have emerged as a powerful paradigm for automating scientific discovery. To differentiate agent behavior in the multi-agent system, current frameworks typically assign generic role-based personas such as ''reviewer'' or ''writer'' or rely on coarse grained keyword-based personas. While functional, this approach oversimplifies how human scientists operate, whose contributions are shaped by their unique research trajectories. In response, we propose INDIBATOR, a framework for molecular discovery that grounds agents in individualized scientist profiles constructed from two modalities: publication history for literature-derived knowledge and molecular history for structural priors. These agents engage in multi-turn debate through proposal, critique, and voting phases. Our evaluation demonstrates that these fine-grained individuality-grounded agents consistently outperform systems relying on coarse-grained personas, achieving competitive or state-of-the-art performance. These results validate that capturing the ``scientific DNA'' of individual agents is essential for high-quality discovery.

Figures (5)

• Figure 1: Overview of $\textsc{Indibator}$. Given a task, the supervisor agent selects relevant scientists by identifying the authors of publications by RAG. Next, individuality is grounded for each agent with scientist profiles, consisting of publication history and molecular history of each scientist. Finally, multi-agents debate to iteratively generate candidate molecules with proposal, critique, and voting phases.
• Figure 2: Results of protein target molecular generation. The left and right panels illustrate the docking scores and diversity of molecules, respectively. The gray, red, teal colors denote vanilla debate, keyword persona debate, and $\textsc{Indibator}$ (ours), respectively. Notably, the docking scores are presented in absolute values, with higher scores representing superior binding.
• Figure 3: Qualitative case study on individuality grounded agents. We provide a qualitative analysis of the JNK3 inhibition guided molecule generation task. Specifically, we show how an agent leverages prior publications and molecules to propose a candidate, while other agents utilize their profiles to offer targeted critiques.
• Figure 4: Effect of the number of collaborators. Annotated numbers above and below each data point indicate the number of debate rounds required to generate 1,000 candidates.
• Figure 5: Fixed total number of proposals.