# Decoupling Reasoning and Knowledge Injection for In-Context Knowledge Editing

arXiv: https://arxiv.org/abs/2506.00536 · HTML: https://mlsanity.com/paper/2506.00536/

## Key facts

- Cited by 12 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 12 of its 37 references that are in the map.
- First posted on arXiv 31 May 2025.

## Placed in

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

## Builds on (most cited first, 12 of 12)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models](https://mlsanity.com/paper/2410.02355/) (2024)
- [jina-embeddings-v3: Multilingual Embeddings With Task LoRA](https://mlsanity.com/paper/2409.10173/) (2024)
- [Parametric Retrieval Augmented Generation](https://mlsanity.com/paper/2501.15915/) (2025)
- [DeepEdit: Knowledge Editing as Decoding with Constraints](https://mlsanity.com/paper/2401.10471/) (2024)
- [RbFT: Robust Fine-tuning for Retrieval-Augmented Generation against Retrieval Defects](https://mlsanity.com/paper/2501.18365/) (2025)
- [Knowledge Editing through Chain-of-Thought](https://mlsanity.com/paper/2412.17727/) (2024)
- [Should We Really Edit Language Models? On the Evaluation of Edited Language Models](https://mlsanity.com/paper/2410.18785/) (2024)
- [JuDGE: Benchmarking Judgment Document Generation for Chinese Legal System](https://mlsanity.com/paper/2503.14258/) (2025)

## Built on by (most cited first, 12 of 12)

- [Skill Retrieval Augmentation for Agentic AI](https://mlsanity.com/paper/2604.24594/) (2026)
- [SurGE: A Benchmark and Evaluation Framework for Scientific Survey Generation](https://mlsanity.com/paper/2508.15658/) (2025)
- [Enhancing Judgment Document Generation via Agentic Legal Information Collection and Rubric-Guided Optimization](https://mlsanity.com/paper/2605.02011/) (2026)
- [Demystifying Scientific Problem-Solving in LLMs by Probing Knowledge and Reasoning](https://mlsanity.com/paper/2508.19202/) (2025)
- [Dynamic and Parametric Retrieval-Augmented Generation](https://mlsanity.com/paper/2506.06704/) (2025)
- [Towards Unification of Hallucination Detection and Fact Verification for Large Language Models](https://mlsanity.com/paper/2512.02772/) (2025)
- [BetaEdit: Null-Space Constrained Sequential Model Editing](https://mlsanity.com/paper/2605.09285/) (2026)
- [Dual-Layer Agentic Memory with Fast Write Routing and Slow Consolidation](https://mlsanity.com/paper/2608.22215/) (2026)
- [Decoupled Mixture-of-Experts for Parametric Knowledge Injection](https://mlsanity.com/paper/2606.14243/) (2026)
- [Retrievable Gradients: Continual Post-Training Without Cumulative Weight Drift](https://mlsanity.com/paper/2606.15734/) (2026)
- [Orthogonal Representation Editing: Decoupling Semantic Entanglement in Batch Knowledge Editing of LLMs](https://mlsanity.com/paper/2606.22627/) (2026)
- [RWGBench: Evaluating Scholarly Positioning in Related Work Generation](https://mlsanity.com/paper/2606.24894/) (2026)

## Abstract (arXiv)

Knowledge editing aims to efficiently update Large Language Models (LLMs) by modifying specific knowledge without retraining the entire model. Among knowledge editing approaches, in-context editing (ICE) offers a lightweight solution by injecting new knowledge directly into the input context, leaving model parameters unchanged. However, existing ICE approaches do not explicitly separate the newly injected knowledge from the model's original reasoning process. This entanglement often results in conflicts between external updates and internal parametric knowledge, undermining the consistency and accuracy of the reasoning path.In this work, we conduct preliminary experiments to examine how parametric knowledge influences reasoning path planning. We find that the model's reasoning is tightly coupled with its internal knowledge, and that naively injecting new information without adapting the reasoning path often leads to performance degradation, particularly in multi-hop tasks. To this end, we propose DecKER, a novel ICE framework that decouples reasoning from knowledge editing by generating a masked reasoning path and then resolving knowledge edits via hybrid retrieval and model-based validation. Experiments on multi-hop QA benchmarks show that DecKER significantly outperforms existing ICE methods by mitigating knowledge conflicts and preserving reasoning consistency. Our code is available at: https://github.com/bebr2/DecKER .