# Investigating Multi-Hop Factual Shortcuts in Knowledge Editing of Large Language Models

arXiv: https://arxiv.org/abs/2402.11900 · HTML: https://mlsanity.com/paper/2402.11900/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning.
- Builds on 25 of its 36 references that are in the map.
- First posted on arXiv 19 Feb 2024.

## Placed in

- LLM reasoning / [Knowledge- and tool-augmented reasoning](https://mlsanity.com/lr/knowledge-and-tool-augmented-reasoning/)

## Builds on (most cited first, 25 of 25)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [A Survey of Large Language Models](https://mlsanity.com/paper/2303.18223/) (2023)
- [Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling](https://mlsanity.com/paper/2304.01373/) (2023)
- [Calibrate Before Use: Improving Few-Shot Performance of Language Models](https://mlsanity.com/paper/2102.09690/) (2021)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Measuring and Narrowing the Compositionality Gap in Language Models](https://mlsanity.com/paper/2210.03350/) (2022)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [Reasoning with Language Model is Planning with World Model](https://mlsanity.com/paper/2305.14992/) (2023)
- [Towards Reasoning in Large Language Models: A Survey](https://mlsanity.com/paper/2212.10403/) (2022)
- [Memory-Based Model Editing at Scale](https://mlsanity.com/paper/2206.06520/) (2022)
- [Zero-Shot Relation Extraction via Reading Comprehension](https://mlsanity.com/paper/1706.04115/) (2017)
- [MQuAKE: Assessing Knowledge Editing in Language Models via Multi-Hop Questions](https://mlsanity.com/paper/2305.14795/) (2023)
- [Can We Edit Factual Knowledge by In-Context Learning?](https://mlsanity.com/paper/2305.12740/) (2023)
- [Transformer-Patcher: One Mistake worth One Neuron](https://mlsanity.com/paper/2301.09785/) (2023)
- [Knowledge Editing for Large Language Models: A Survey](https://mlsanity.com/paper/2310.16218/) (2023)
- [Navigate through Enigmatic Labyrinth A Survey of Chain of Thought Reasoning: Advances, Frontiers and Future](https://mlsanity.com/paper/2309.15402/) (2023)
- [Recall and Learn: Fine-tuning Deep Pretrained Language Models with Less Forgetting](https://mlsanity.com/paper/2004.12651/) (2020)
- [Calibrating Factual Knowledge in Pretrained Language Models](https://mlsanity.com/paper/2210.03329/) (2022)
- [Improving In-Context Few-Shot Learning via Self-Supervised Training](https://mlsanity.com/paper/2205.01703/) (2022)
- [Plug-and-Play Adaptation for Continuously-updated QA](https://mlsanity.com/paper/2204.12785/) (2022)

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

- [Knowledge Circuits in Pretrained Transformers](https://mlsanity.com/paper/2405.17969/) (2024)
- [SAKURA: On the Multi-hop Reasoning of Large Audio-Language Models Based on Speech and Audio Information](https://mlsanity.com/paper/2505.13237/) (2025)
- [Locate-then-edit for Multi-hop Factual Recall under Knowledge Editing](https://mlsanity.com/paper/2410.06331/) (2024)
- [Do Large Language Models Perform Latent Multi-Hop Reasoning without Exploiting Shortcuts?](https://mlsanity.com/paper/2411.16679/) (2024)
- [Back Attention: Understanding and Enhancing Multi-Hop Reasoning in Large Language Models](https://mlsanity.com/paper/2502.10835/) (2025)
- [Co-occurrence is not Factual Association in Language Models](https://mlsanity.com/paper/2409.14057/) (2024)
- [Do Latent-CoT Models Think Step-by-Step? A Mechanistic Study on Sequential Reasoning Tasks](https://mlsanity.com/paper/2602.00449/) (2026)
- [Compositional Generalization from Learned Skills via CoT Training: A Theoretical and Structural Analysis for Reasoning](https://mlsanity.com/paper/2502.04667/) (2025)
- [Beyond Scaling: Measuring and Predicting the Upper Bound of Knowledge Retention in Language Model Pre-Training](https://mlsanity.com/paper/2502.04066/) (2025)
- [Can Memory-Augmented Language Models Generalize on Reasoning-in-a-Haystack Tasks?](https://mlsanity.com/paper/2503.07903/) (2025)
- [EAMET: Robust Massive Model Editing via Embedding Alignment Optimization](https://mlsanity.com/paper/2505.11876/) (2025)
- [RADAR: Reasoning as Discrimination with Aligned Representations for LLM-based Knowledge Graph Reasoning](https://mlsanity.com/paper/2602.21951/) (2026)
- [The Power of Power Law: Asymmetry Enables Compositional Reasoning](https://mlsanity.com/paper/2604.22951/) (2026)
- [Synthetic Worlds for Temporal Evaluation and Knowledge Updating in LLMs](https://mlsanity.com/paper/2609.00184/) (2026)

## Abstract (arXiv)

Recent work has showcased the powerful capability of large language models (LLMs) in recalling knowledge and reasoning. However, the reliability of LLMs in combining these two capabilities into reasoning through multi-hop facts has not been widely explored. This paper systematically investigates the possibilities for LLMs to utilize shortcuts based on direct connections between the initial and terminal entities of multi-hop knowledge. We first explore the existence of factual shortcuts through Knowledge Neurons, revealing that: (i) the strength of factual shortcuts is highly correlated with the frequency of co-occurrence of initial and terminal entities in the pre-training corpora; (ii) few-shot prompting leverage more shortcuts in answering multi-hop questions compared to chain-of-thought prompting. Then, we analyze the risks posed by factual shortcuts from the perspective of multi-hop knowledge editing. Analysis shows that approximately 20% of the failures are attributed to shortcuts, and the initial and terminal entities in these failure instances usually have higher co-occurrences in the pre-training corpus. Finally, we propose erasing shortcut neurons to mitigate the associated risks and find that this approach significantly reduces failures in multiple-hop knowledge editing caused by shortcuts.