# Learning When Not to Answer: A Ternary Reward Structure for Reinforcement Learning based Question Answering

arXiv: https://arxiv.org/abs/1902.10236 · HTML: https://mlsanity.com/paper/1902.10236/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Feb 2019.

## Built on by (most cited first, 7 of 7)

- [Reinforcement Learning from Reformulations in Conversational Question Answering over Knowledge Graphs](https://mlsanity.com/paper/2105.04850/) (2021)
- [Nano: Nested Human-in-the-Loop Reward Learning for Few-shot Language Model Control](https://mlsanity.com/paper/2211.05750/) (2022)
- [Ranking vs. Classifying: Measuring Knowledge Base Completion Quality](https://mlsanity.com/paper/2102.06145/) (2021)
- [Towards Reliable and Factual Response Generation: Detecting Unanswerable Questions in Information-Seeking Conversations](https://mlsanity.com/paper/2401.11452/) (2024)
- [Rewarding Intellectual Humility Learning When Not To Answer In Large Language Models](https://mlsanity.com/paper/2601.20126/) (2026)
- [Reflect-R1: Evidence-Driven Reflection for Self-Correction in Long Video Understanding](https://mlsanity.com/paper/2606.27922/) (2026)
- [ReInform: Selecting paths with reinforcement learning for contextualized link prediction](https://mlsanity.com/paper/2211.10688/) (2022)

## Abstract (arXiv)

In this paper, we investigate the challenges of using reinforcement learning agents for question-answering over knowledge graphs for real-world applications. We examine the performance metrics used by state-of-the-art systems and determine that they are inadequate for such settings. More specifically, they do not evaluate the systems correctly for situations when there is no answer available and thus agents optimized for these metrics are poor at modeling confidence. We introduce a simple new performance metric for evaluating question-answering agents that is more representative of practical usage conditions, and optimize for this metric by extending the binary reward structure used in prior work to a ternary reward structure which also rewards an agent for not answering a question rather than giving an incorrect answer. We show that this can drastically improve the precision of answered questions while only not answering a limited number of previously correctly answered questions. Employing a supervised learning strategy using depth-first-search paths to bootstrap the reinforcement learning algorithm further improves performance.