DualReward: A Dynamic Reinforcement Learning Framework for Cloze Tests Distractor Generation
This work addresses the problem of generating high-quality distractors for automated test creation, offering a flexible framework that balances learning from human examples and exploring novel distractors, though it appears incremental in its approach.
The paper tackled automatic distractor generation for cloze tests by introducing DualReward, a reinforcement learning framework with adaptive reward scaling, resulting in modest improvements on homogeneous datasets and substantial gains of 3.48-3.86% in P@1 on diverse cross-domain data.
This paper introduces DualReward, a novel reinforcement learning framework for automatic distractor generation in cloze tests. Unlike conventional approaches that rely primarily on supervised learning or static generative models, our method employs a dual reward structure with adaptive scaling that differentiates between human-created gold standard distractors and model-generated candidates. The framework dynamically adjusts reward signal intensity based on model performance and confidence. We evaluate our approach on both passage-level (CLOTH-F) and sentence-level (MCQ) cloze test datasets, demonstrating consistent improvements over state-of-the-art baselines. Experimental results show that our adaptive reward scaling mechanism provides modest but consistent benefits on homogeneous datasets (CLOTH-F) and more substantial improvements (3.48-3.86% in P@1) on diverse, cross-domain data (MCQ), suggesting its particular effectiveness for handling varied question types and domains. Our work offers a flexible framework that effectively balances learning from reliable human examples while exploring novel, high-quality distractors for automated test generation.