Quality-diverse recommendations improve decision-making with reasons included
QD-Rec: Open-Ended Evolution of Recommendations via Quality-Diversity
Information RetrievalArtificial IntelligenceMachine LearningNeural and Evolutionary Computing
Summary
Choosing among options can be hard, especially when the best choice depends on detailed reasons rather than just scores. The authors introduce a way to suggest a variety of good choices, each with an explanation of why it fits the decision. Their method, called QD-Rec, uses a mix of exploration and a language model to keep improving and diversifying recommendations. This makes it easier for decision makers to compare options with clear, diverse reasons.
What this means in practice
- •For mergers and acquisitions teams: Recommend diverse companies to acquire, each with a rationale tailored to strategic needs and risks.
- •For marketing analysts: Generate varied customer targeting options with natural language reasons, improving targeted campaigns.
Authors
Takashi Ishida, Taro Makino, Issa Sugiura
Abstract
Open-ended search combined with foundation models has produced advances such as faster algorithms, new results on open mathematical problems, and self-improving agents. We extend this paradigm to recommendation for decision-making, focusing on decisions where the set of possible choices is fixed but the reasons for choosing them are not. In high-stakes decisions, e.g., which company to acquire or which customer to target, comparing diverse, well-reasoned options can be more useful than a single ranked list. We formalize open-ended recommendation as building a diverse, high-quality set of recommendations, each an item together with a hypothesis, a natural-language reason why the item fits the needs of a decision maker. We propose QD-Rec, a quality--diversity method that selects items with a contextual bandit and generates hypotheses with a foundation model from earlier ones kept in an archive as stepping stones. A new recommendation enters the archive if it is novel or better than its nearest neighbor. Across M&A, customer targeting, and paper and book recommendation, QD-Rec gives the highest mean quality and builds the largest archive in most settings.