# HyperPatch: Sequential Knowledge Editing Under n-ary Structural Drift

arXiv: https://arxiv.org/abs/2606.03179 · HTML: https://mlsanity.com/paper/2606.03179/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 41 references that are in the map.
- First posted on arXiv 2 Jun 2026.

## Builds on (most cited first, 22 of 22)

- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [From Local to Global: A Graph RAG Approach to Query-Focused Summarization](https://mlsanity.com/paper/2404.16130/) (2024)
- [BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models](https://mlsanity.com/paper/2104.08663/) (2021)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [Hypergraph Neural Networks](https://mlsanity.com/paper/1809.09401/) (2018)
- [The Faiss library](https://mlsanity.com/paper/2401.08281/) (2024)
- [Editing Large Language Models: Problems, Methods, and Opportunities](https://mlsanity.com/paper/2305.13172/) (2023)
- [MQuAKE: Assessing Knowledge Editing in Language Models via Multi-Hop Questions](https://mlsanity.com/paper/2305.14795/) (2023)
- [Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling](https://mlsanity.com/paper/2010.11304/) (2020)
- [Compositional Questions Do Not Necessitate Multi-hop Reasoning](https://mlsanity.com/paper/1906.02900/) (2019)
- [Memory Augmented Large Language Models are Computationally Universal](https://mlsanity.com/paper/2301.04589/) (2023)
- [PokeMQA: Programmable knowledge editing for Multi-hop Question Answering](https://mlsanity.com/paper/2312.15194/) (2023)
- [GraphLoRA: Structure-Aware Contrastive Low-Rank Adaptation for Cross-Graph Transfer Learning](https://mlsanity.com/paper/2409.16670/) (2024)
- [Question Decomposition for Retrieval-Augmented Generation](https://mlsanity.com/paper/2507.00355/) (2025)
- [Knowledge Editing with Dynamic Knowledge Graphs for Multi-Hop Question Answering](https://mlsanity.com/paper/2412.13782/) (2024)
- [Tailoring Table Retrieval from a Field-aware Hybrid Matching Perspective](https://mlsanity.com/paper/2503.02251/) (2025)
- [Is Model Editing Built on Sand? Revealing Its Illusory Success and Fragile Foundation](https://mlsanity.com/paper/2510.00625/) (2025)
- [Decoupling Spatio-Temporal Prediction: When Lightweight Large Models Meet Adaptive Hypergraphs](https://mlsanity.com/paper/2505.19620/) (2025)
- [Are We Evaluating the Edit Locality of LLM Model Editing Properly?](https://mlsanity.com/paper/2601.17343/) (2026)
- [Modeling Edge-Specific Node Features through Co-Representation Neural Hypergraph Diffusion](https://mlsanity.com/paper/2405.14286/) (2024)

## Abstract (arXiv)

Large Language Models (LLMs) rely on Knowledge Editing (KE) to maintain temporal validity, yet real-world knowledge is inherently n-ary. We demonstrate that in non-stationary environments, sequential updates to complex relations induce N-ary Structural Drift, a phenomenon where the binary reification of n-ary events into triples fractures relational atomicity. This precipitates Structure-Conditioned Knowledge Transfer Failure, a systematic mis-grounding of the retriever frequently misdiagnosed as parametric hallucination. To tackle this, we propose HyperPatch, a parameter-preserving framework that reformulates sequential KE as a stability problem over hypergraph manifolds. HyperPatch preserves event integrity through three phases: (i) Structural Prior Initialization, establishing a topology-aware embedding space via contrastive learning on a Hypergraph Neural Network (HGNN) to capture high-order correlations; (ii) Sequential Topology Editing, utilizing a dual-stage mechanism that employs SimHash-based Topological Alignment for rapid conflict resolution and Topological LoRA Adaptation to track drift without backbone retraining; and (iii) Structure-Conditioned Reasoning, which integrates globally consistent evidence from fused linguistic and structural manifolds. On the MQuAKE-CF and MQuAKE-T benchmarks, HyperPatch achieves relative gains in Hop-wise Accuracy (H-Acc) of 96.24% and 21.06% over the strongest baseline, respectively. Further ablations demonstrate superior reliability under continuous n-ary update streams, whereas the standard KG-based variant suffers H-Acc collapses of up to 88.3% due to structural misalignment.