# Is Graph Structure Necessary for Multi-hop Question Answering?

arXiv: https://arxiv.org/abs/2004.03096 · HTML: https://mlsanity.com/paper/2004.03096/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Apr 2020.

## Built on by (most cited first, 6 of 6)

- [Locate Then Ask: Interpretable Stepwise Reasoning for Multi-hop Question Answering](https://mlsanity.com/paper/2208.10297/) (2022)
- [Memory Augmented Sequential Paragraph Retrieval for Multi-hop Question Answering](https://mlsanity.com/paper/2102.03741/) (2021)
- [Reasoning Chain Based Adversarial Attack for Multi-hop Question Answering](https://mlsanity.com/paper/2112.09658/) (2021)
- [Towards Interpretable and Reliable Reading Comprehension: A Pipeline Model with Unanswerability Prediction](https://mlsanity.com/paper/2111.09029/) (2021)
- [Generative Context Pair Selection for Multi-hop Question Answering](https://mlsanity.com/paper/2104.08744/) (2021)
- [Parameter-Efficient Neural Question Answering Models via Graph-Enriched Document Representations](https://mlsanity.com/paper/2106.00851/) (2021)

## Abstract (arXiv)

Recently, attempting to model texts as graph structure and introducing graph neural networks to deal with it has become a trend in many NLP research areas. In this paper, we investigate whether the graph structure is necessary for multi-hop question answering. Our analysis is centered on HotpotQA. We construct a strong baseline model to establish that, with the proper use of pre-trained models, graph structure may not be necessary for multi-hop question answering. We point out that both graph structure and adjacency matrix are task-related prior knowledge, and graph-attention can be considered as a special case of self-attention. Experiments and visualized analysis demonstrate that graph-attention or the entire graph structure can be replaced by self-attention or Transformers.