An Answer is just the Start: Related Insight Generation for Open-Ended Document-Grounded QA
For researchers and practitioners in open-ended QA, this work addresses the lack of benchmarks and methods for generating related insights that help users refine answers, though the task is domain-specific to scientific collections.
The paper introduces a new task, document-grounded related insight generation, to support iterative refinement of answers to open-ended questions. The authors curate the SCOpE-QA dataset and propose InsightGen, which achieves consistent improvements in generating useful, relevant, and actionable insights over baselines.
Answering open-ended questions remains challenging for AI systems because it requires synthesis, judgment, and exploration beyond factual retrieval, and users often refine answers through multiple iterations rather than accepting a single response. Existing QA benchmarks do not explicitly support this refinement process. To address this gap, we introduce a new task, document-grounded related insight generation, where the goal is to generate additional insights from a document collection that help improve, extend, or rethink an initial answer to an open-ended question, ultimately supporting richer user interaction and a better overall question answering experience. We curate and release SCOpE-QA (Scientific Collections for Open-Ended QA), a dataset of 3,000 open-ended questions across 20 research collections. We present InsightGen, a two-stage approach that first constructs a thematic representation of the document collection using clustering, and then selects related context based on neighborhood selection from the thematic graph to generate diverse and relevant insights using LLMs. Extensive evaluation on 3,000 questions using two generation models and two evaluation settings shows that InsightGen consistently produces useful, relevant, and actionable insights, establishing a strong baseline for this new task.