WikiHint: A Human-Annotated Dataset for Hint Ranking and Generation

TL;DR

This paper introduces WikiHint, a human-annotated dataset for hint ranking and generation, and demonstrates its effectiveness in improving hint quality and ranking accuracy using fine-tuned LLMs and a new evaluation method.

Contribution

The paper presents WikiHint, a new dataset for hint generation and ranking, along with fine-tuning methods and a lightweight evaluation metric for improved hint quality.

Findings

Hints generated with the dataset are more effective.

Including answer information improves hint quality.

Encoder-based models outperform decoder-based models in ranking.

Abstract

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.