# Editing Common Sense in Transformers

arXiv: https://arxiv.org/abs/2305.14956 · HTML: https://mlsanity.com/paper/2305.14956/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 45 references that are in the map.
- First posted on arXiv 24 May 2023.

## Builds on (most cited first, 29 of 29)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [GPT-NeoX-20B: An Open-Source Autoregressive Language Model](https://mlsanity.com/paper/2204.06745/) (2022)
- [Retrieval Augmentation Reduces Hallucination in Conversation](https://mlsanity.com/paper/2104.07567/) (2021)
- [Knowledge Neurons in Pretrained Transformers](https://mlsanity.com/paper/2104.08696/) (2021)
- [Editing Factual Knowledge in Language Models](https://mlsanity.com/paper/2104.08164/) (2021)
- [Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space](https://mlsanity.com/paper/2203.14680/) (2022)
- [Fast Model Editing at Scale](https://mlsanity.com/paper/2110.11309/) (2021)
- [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)
- [Direct and Indirect Effects](https://mlsanity.com/paper/1301.2300/) (2013)
- [Surface Form Competition: Why the Highest Probability Answer Isn't Always Right](https://mlsanity.com/paper/2104.08315/) (2021)
- [Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models](https://mlsanity.com/paper/2301.04213/) (2023)
- [Modifying Memories in Transformer Models](https://mlsanity.com/paper/2012.00363/) (2020)
- [Editable Neural Networks](https://mlsanity.com/paper/2004.00345/) (2020)
- [Rewriting a Deep Generative Model](https://mlsanity.com/paper/2007.15646/) (2020)
- [Evaluating the Factual Consistency of Large Language Models Through News Summarization](https://mlsanity.com/paper/2211.08412/) (2022)
- [Can LMs Learn New Entities from Descriptions? Challenges in Propagating Injected Knowledge](https://mlsanity.com/paper/2305.01651/) (2023)
- [Gradient-based Editing of Memory Examples for Online Task-free Continual Learning](https://mlsanity.com/paper/2006.15294/) (2020)
- [Commonsense Knowledge Reasoning and Generation with Pre-trained Language Models: A Survey](https://mlsanity.com/paper/2201.12438/) (2022)
- [Kformer: Knowledge Injection in Transformer Feed-Forward Layers](https://mlsanity.com/paper/2201.05742/) (2022)
- [Learning to Repair: Repairing model output errors after deployment using a dynamic memory of feedback](https://mlsanity.com/paper/2112.09737/) (2021)
- [Evaluating Factuality in Text Simplification](https://mlsanity.com/paper/2204.07562/) (2022)
- [Modeling Semantic Plausibility by Injecting World Knowledge](https://mlsanity.com/paper/1804.00619/) (2018)
- [Modeling Event Plausibility with Consistent Conceptual Abstraction](https://mlsanity.com/paper/2104.10247/) (2021)

## Built on by (most cited first, 28 of 28)

- [Rethinking Machine Unlearning for Large Language Models](https://mlsanity.com/paper/2402.08787/) (2024)
- [Knowledge Editing for Large Language Models: A Survey](https://mlsanity.com/paper/2310.16218/) (2023)
- [A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models](https://mlsanity.com/paper/2407.02646/) (2024)
- [A Survey on Knowledge Editing of Neural Networks](https://mlsanity.com/paper/2310.19704/) (2023)
- [In-Context Editing: Learning Knowledge from Self-Induced Distributions](https://mlsanity.com/paper/2406.11194/) (2024)
- [Fundamental Problems With Model Editing: How Should Rational Belief Revision Work in LLMs?](https://mlsanity.com/paper/2406.19354/) (2024)
- [LLM-Based Multi-Hop Question Answering with Knowledge Graph Integration in Evolving Environments](https://mlsanity.com/paper/2408.15903/) (2024)
- [Should We Really Edit Language Models? On the Evaluation of Edited Language Models](https://mlsanity.com/paper/2410.18785/) (2024)
- [Discovering Knowledge-Critical Subnetworks in Pretrained Language Models](https://mlsanity.com/paper/2310.03084/) (2023)
- [Commonsense Knowledge Editing Based on Free-Text in LLMs](https://mlsanity.com/paper/2410.23844/) (2024)
- [A Causal Lens for Evaluating Faithfulness Metrics](https://mlsanity.com/paper/2502.18848/) (2025)
- [Concept-ROT: Poisoning Concepts in Large Language Models with Model Editing](https://mlsanity.com/paper/2412.13341/) (2024)
- [Bidirectional LMs are Better Knowledge Memorizers? A Benchmark for Real-world Knowledge Injection](https://mlsanity.com/paper/2505.12306/) (2025)
- [Towards Practical Tool Usage for Continually Learning LLMs](https://mlsanity.com/paper/2404.09339/) (2024)
- [Continuously Steering LLMs Sensitivity to Contextual Knowledge with Proxy Models](https://mlsanity.com/paper/2508.19720/) (2025)
- [How Well Can Knowledge Edit Methods Edit Perplexing Knowledge?](https://mlsanity.com/paper/2406.17253/) (2024)
- [Learning Where to Edit Vision Transformers](https://mlsanity.com/paper/2411.01948/) (2024)
- [From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers](https://mlsanity.com/paper/2506.17052/) (2025)
- [Locating and Extracting Relational Concepts in Large Language Models](https://mlsanity.com/paper/2406.13184/) (2024)
- [FAME: Towards Factual Multi-Task Model Editing](https://mlsanity.com/paper/2410.10859/) (2024)
- [SetKE: Knowledge Editing for Knowledge Elements Overlap](https://mlsanity.com/paper/2504.20972/) (2025)
- [AnyEdit++: Adaptive Long-Form Knowledge Editing via Bayesian Surprise](https://mlsanity.com/paper/2606.01053/) (2026)
- [Expanding before Inferring: Enhancing Factuality in Large Language Models through Premature Layers Interpolation](https://mlsanity.com/paper/2506.02973/) (2025)
- [Know-MRI: A Knowledge Mechanisms Revealer&Interpreter for Large Language Models](https://mlsanity.com/paper/2506.08427/) (2025)
- [Knowledge Editing for Multi-Hop Question Answering Using Semantic Analysis](https://mlsanity.com/paper/2508.00914/) (2025)
- [Lower Layers Matter: Alleviating Hallucination via Multi-Layer Fusion Contrastive Decoding with Truthfulness Refocused](https://mlsanity.com/paper/2408.08769/) (2024)
- [Break Through the Compression Bottleneck: From Theory to Practice](https://mlsanity.com/paper/2607.20434/) (2026)
- [ManiEdit: Sequential Unstructured Knowledge Editing for Language Models from a Manifold Perspective](https://mlsanity.com/paper/2609.33534/) (2026)

## Abstract (arXiv)

Editing model parameters directly in Transformers makes updating open-source transformer-based models possible without re-training (Meng et al., 2023). However, these editing methods have only been evaluated on statements about encyclopedic knowledge with a single correct answer. Commonsense knowledge with multiple correct answers, e.g., an apple can be green or red but not transparent, has not been studied but is as essential for enhancing transformers' reliability and usefulness. In this paper, we investigate whether commonsense judgments are causally associated with localized, editable parameters in Transformers, and we provide an affirmative answer. We find that directly applying the MEMIT editing algorithm results in sub-par performance and improve it for the commonsense domain by varying edit tokens and improving the layer selection strategy, i.e., MEMITCSK. GPT-2 Large and XL models edited using MEMITCSK outperform best-fine-tuned baselines by 10.97% and 10.73% F1 scores on PEP3k and 20Q datasets. In addition, we propose a novel evaluation dataset, PROBE SET, that contains unaffected and affected neighborhoods, affected paraphrases, and affected reasoning challenges. MEMITCSK performs well across the metrics while fine-tuning baselines show significant trade-offs between unaffected and affected metrics. These results suggest a compelling future direction for incorporating feedback about common sense into Transformers through direct model editing.