# Do Multi-hop Readers Dream of Reasoning Chains?

arXiv: https://arxiv.org/abs/1910.14520 · HTML: https://mlsanity.com/paper/1910.14520/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Oct 2019.

## Built on by (most cited first, 14 of 14)

- [Exploring Length Generalization in Large Language Models](https://mlsanity.com/paper/2207.04901/) (2022)
- [Teaching Algorithmic Reasoning via In-context Learning](https://mlsanity.com/paper/2211.09066/) (2022)
- [Baleen: Robust Multi-Hop Reasoning at Scale via Condensed Retrieval](https://mlsanity.com/paper/2101.00436/) (2021)
- [From LSAT: The Progress and Challenges of Complex Reasoning](https://mlsanity.com/paper/2108.00648/) (2021)
- [Reasoning Abilities of Large Language Models: In-Depth Analysis on the Abstraction and Reasoning Corpus](https://mlsanity.com/paper/2403.11793/) (2024)
- [Adaptive Information Seeking for Open-Domain Question Answering](https://mlsanity.com/paper/2109.06747/) (2021)
- [WEC: Deriving a Large-scale Cross-document Event Coreference dataset from Wikipedia](https://mlsanity.com/paper/2104.05022/) (2021)
- [SciCo: Hierarchical Cross-Document Coreference for Scientific Concepts](https://mlsanity.com/paper/2104.08809/) (2021)
- [Hopper: Multi-hop Transformer for Spatiotemporal Reasoning](https://mlsanity.com/paper/2103.10574/) (2021)
- [Cross-document Event Coreference Search: Task, Dataset and Modeling](https://mlsanity.com/paper/2210.12654/) (2022)
- [Complementary Evidence Identification in Open-Domain Question Answering](https://mlsanity.com/paper/2103.11643/) (2021)
- [Bactrainus: Optimizing Large Language Models for Multi-hop Complex Question Answering Tasks](https://mlsanity.com/paper/2501.06286/) (2025)
- [Feudal Reinforcement Learning by Reading Manuals](https://mlsanity.com/paper/2110.06477/) (2021)
- [JECC: Commonsense Reasoning Tasks Derived from Interactive Fictions](https://mlsanity.com/paper/2210.15456/) (2022)

## Abstract (arXiv)

General Question Answering (QA) systems over texts require the multi-hop reasoning capability, i.e. the ability to reason with information collected from multiple passages to derive the answer. In this paper we conduct a systematic analysis to assess such an ability of various existing models proposed for multi-hop QA tasks. Specifically, our analysis investigates that whether providing the full reasoning chain of multiple passages, instead of just one final passage where the answer appears, could improve the performance of the existing QA models. Surprisingly, when using the additional evidence passages, the improvements of all the existing multi-hop reading approaches are rather limited, with the highest error reduction of 5.8% on F1 (corresponding to 1.3% absolute improvement) from the BERT model. To better understand whether the reasoning chains could indeed help find correct answers, we further develop a co-matching-based method that leads to 13.1% error reduction with passage chains when applied to two of our base readers (including BERT). Our results demonstrate the existence of the potential improvement using explicit multi-hop reasoning and the necessity to develop models with better reasoning abilities.