WikiHint: A Human-Annotated Dataset for Hint Ranking and Generation
This work addresses the problem of maintaining reasoning skills in users of AI chatbots by providing a dataset and methods for hint generation and ranking, though it is incremental as it builds on existing LLM techniques.
The paper tackles the challenge of preserving human cognitive abilities by promoting hints over direct answers, introducing WikiHint, a manually constructed dataset of 5,000 hints for 1,000 questions, and showing that finetuned LLMs generate more effective hints, with answer-aware contexts improving quality and encoder-based models outperforming in ranking.
The use of Large Language Models (LLMs) has increased significantly with users frequently asking questions to chatbots. In the time when information is readily accessible, it is crucial to stimulate and preserve human cognitive abilities and maintain strong reasoning skills. This paper addresses such challenges by promoting the use of hints as an alternative or a supplement to direct answers. We first introduce a manually constructed hint dataset, WikiHint, which is based on Wikipedia and includes 5,000 hints created for 1,000 questions. We then finetune open-source LLMs for hint generation in answer-aware and answer-agnostic contexts. We assess the effectiveness of the hints with human participants who answer questions with and without the aid of hints. Additionally, we introduce a lightweight evaluation method, HintRank, to evaluate and rank hints in both answer-aware and answer-agnostic settings. Our findings show that (a) the dataset helps generate more effective hints, (b) including answer information along with questions generally improves the quality of generated hints, and (c) encoder-based models perform better than decoder-based models in hint ranking.