Iterative Research Conversations

A bounded question-answer loop lets retrieved evidence expose gaps and drive deeper follow-up searches during agentic research.

Model research as a conversation between a question-asking writer and a source-grounded expert. Each answer becomes context for the next question, so newly discovered details can reveal missing evidence and induce deeper follow-ups. Decompose complex questions into search queries, filter results for source quality, synthesize only supported answers, and accumulate the retained references.

Bound both the number of perspective-guided branches and rounds per branch. This makes discovery depth and cost explicit while avoiding a single static query's dependence on what the agent knew before research began.

---

Sources

• Shao et al. (2024), full paper, pp. 1–27