An Answer is just the Start: Related Insight Generation for Open-Ended Document-Grounded QA
2026-04-21 • Computation and Language
Computation and Language
AI summaryⓘ
The authors point out that current AI systems struggle with open-ended questions because they need to go beyond just finding facts and often require several rounds of refinement. To help with this, they created a new task where AI generates extra helpful insights from a set of documents to improve or rethink initial answers. They collected a dataset called SCOpE-QA with 3,000 questions from scientific papers and introduced InsightGen, a method that groups documents by theme and uses these groups to find relevant information for generating insights. Their tests show InsightGen can consistently produce useful and relevant insights, providing a strong starting point for this kind of question answering.
open-ended questionsquestion answeringdocument clusteringthematic representationlarge language modelsdataset SCOpE-QAinsight generationinformation retrievalAI refinementscientific collections
Authors
Saransh Sharma, Pritika Ramu, Aparna Garimella, Koyel Mukherjee
Abstract
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.