IntElicit: Eliciting and Assessing Contextualized Creativity via Dialogue Policy Optimization

2026-06-10

AI Summary

Existing static evaluation methods struggle to accurately assess genuine creativity, as they are often confounded by non-creative factors such as domain knowledge and participant engagement, and are ill-suited for human–AI co-creative settings. To address these limitations, this work proposes a reinforcement learning–based adaptive AI interviewer framework that supports user-driven ideation through multi-turn, non-directive dialogic scaffolding. The framework incorporates a decomposed process reward mechanism designed to encourage the AI to pose reasoning-provoking questions rather than provide direct answers, thereby mitigating sparse rewards and reward hacking. In both simulated experiments and a human study involving 64 participants, the proposed approach significantly outperformed expert-designed baselines, demonstrating a superior capacity to elicit creative potential that is typically overlooked by conventional static assessments.

Abstract

Contextualized assessment offers high ecological validity for evaluating creativity but introduces a critical challenge: observed performance may be confounded with cognitive proficiency (domain knowledge) and agency (willingness to engage). Meanwhile, in the age of generative AI, creative problem solving increasingly occurs in tool-mediated and human--AI interactive environments, making fully static assessment less aligned with contemporary creative practice. To address these issues, this paper proposes IntElicit, a framework for eliciting and assessing contextualized creativity via dialogue policy optimization. IntElicit functions as a constrained adaptive AI Interviewer: it provides non-directive knowledge and agency scaffolds in multi-turn interaction to reduce non-creative confounders, while preserving participants' responsibility for generating the creative content being evaluated. Specifically, to tackle sparse rewards and potential reward hacking (e.g., answer dictation) in open-ended educational dialogue, IntElicit introduces a decomposed process reward mechanism. This mechanism aligns the policy with pedagogical elicitation, rewarding prompts that draw out participant reasoning rather than producing optimal answers on their behalf. Extensive experiments, including participant simulation and a human subject study (N=64), show that IntElicit improves elicited creative outcomes over expert-designed baselines. Together, the results suggest that interactive elicitation can reveal creative potential that static FPSP-style assessment may miss, providing a formative and diagnostic lens for contextualized creativity assessment in AI-mediated learning contexts.

Problem

• contextualized creativity
• creativity assessment
• human-AI interaction
• confounding factors
• ecological validity

Innovation

• dialogue policy optimization
• contextualized creativity assessment
• process reward mechanism
• adaptive AI interviewer
• creative elicitation