mlsanityData to 14 Oct 2026

WikiBigEdit: Understanding the Limits of Lifelong Knowledge Editing in LLMs

arXiv 2503.05683 ↗First posted 7 Mar 2025cs.CL cs.LGcited by 26

Not on the map. Why?

Abstract (arXiv)

Keeping large language models factually up-to-date is crucial for deployment, yet costly retraining remains a challenge. Knowledge editing offers a promising alternative, but methods are only tested on small-scale or synthetic edit benchmarks. In this work, we aim to bridge research into lifelong knowledge editing to real-world edits at a practically relevant scale. We first introduce WikiBigEdit; a large-scale benchmark of real-world Wikidata edits, built to automatically extend lifelong for future-proof benchmarking. In its first instance, it includes over 500K question-answer pairs for knowledge editing alongside a comprehensive evaluation pipeline. Finally, we use WikiBigEdit to study existing knowledge editing techniques' ability to incorporate large volumes of real-world facts and contrast their capabilities to generic modification techniques such as retrieval augmentation and continual finetuning to acquire a complete picture of the practical extent of current lifelong knowledge editing.

Key facts. Cited by 26 papers in the map (data through 14 Oct 2026). Builds on 66 of its 85 references that are in the map. First posted on arXiv 7 Mar 2025.

Builds on

66 of 85 references.

References · 66

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  4. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  5. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  6. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  7. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  8. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
  9. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,141
  10. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,998
  11. 2021Improving language models by retrieving from trillions of tokenscited by 1,293
  12. 2021BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-modelscited by 1,245
38 more
  1. 2024DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Modelcited by 1,135
  2. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,128
  3. 2022Mass-Editing Memory in a Transformercited by 1,001
  4. 2020MPNet: Masked and Permuted Pre-training for Language Understandingcited by 981
  5. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 810
  6. 2021Robust fine-tuning of zero-shot modelscited by 797
  7. 2019ELI5: Long Form Question Answeringcited by 575
  8. 2021Fast Model Editing at Scalecited by 498
  9. 2024GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projectioncited by 463
  10. 2022Memory-Based Model Editing at Scalecited by 447
  11. 2017Zero-Shot Relation Extraction via Reading Comprehensioncited by 426
  12. 2023FreshLLMs: Refreshing Large Language Models with Search Engine Augmentationcited by 318
  13. 2024HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Modelscited by 281
  14. 2022Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptorscited by 271
  15. 2021Measuring Attribution in Natural Language Generation Modelscited by 256
  16. 2024Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunitiescited by 254
  17. 2022Patching open-vocabulary models by interpolating weightscited by 190
  18. 2024Large Language Models for Education: A Survey and Outlookcited by 188
  19. 2024A Comprehensive Study of Knowledge Editing for Large Language Modelscited by 163
  20. 2024Infini-gram: Scaling Unbounded n-gram Language Models to a Trillion Tokenscited by 153
  21. 2024SaulLM-7B: A pioneering Large Language Model for Lawcited by 142
  22. 2023Large Language Models in Law: A Surveycited by 128
  23. 2024Simple and Scalable Strategies to Continually Pre-train Large Language Modelscited by 127
  24. 2022StreamingQA: A Benchmark for Adaptation to New Knowledge over Time in Question Answering Modelscited by 126
  25. 2022TemporalWiki: A Lifelong Benchmark for Training and Evaluating Ever-Evolving Language Modelscited by 119
  26. 2024Model Editing at Scale leads to Gradual and Catastrophic Forgettingcited by 93
  27. 2023DISC-MedLLM: Bridging General Large Language Models and Real-World Medical Consultationcited by 90
  28. 2024No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performancecited by 75
  29. 2024MagMax: Leveraging Model Merging for Seamless Continual Learningcited by 67
  30. 2023Computationally Budgeted Continual Learning: What Does Matter?cited by 61
  31. 2024The Butterfly Effect of Model Editing: Few Edits Can Trigger Large Language Models Collapsecited by 54
  32. 2024Dated Data: Tracing Knowledge Cutoffs in Large Language Modelscited by 51
  33. 2024Rebuilding ROME : Resolving Model Collapse during Sequential Model Editingcited by 50
  34. 2023Weighted Ensemble Models Are Strong Continual Learnerscited by 42
  35. 2024Lifelong Knowledge Editing for LLMs with Retrieval-Augmented Continuous Prompt Learningcited by 40
  36. 2021ArchivalQA: A Large-scale Benchmark Dataset for Open Domain Question Answering over Historical News Collectionscited by 30
  37. 2023Online Continual Learning Without the Storage Constraintcited by 29
  38. 2024Medical Adaptation of Large Language and Vision-Language Models: Are We Making Progress?cited by 28

50 most cited of 66 All 66 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 7; 2026: 19.

26 papers.

Sort:

By area:

  1. 2025UltraEdit: Training-, Subject-, and Memory-Free Lifelong Editing in Language Modelscited by 9
  2. 2025Fine-tuning Done Right in Model Editingcited by 9
  3. 2025Edit Less, Achieve More: Dynamic Sparse Neuron Masking for Lifelong Knowledge Editing in LLMscited by 7
  4. 2026CrispEdit: Low-Curvature Projections for Scalable Non-Destructive LLM Editingcited by 5
  5. 2026RAG or Learning? Understanding the Limits of LLM Adaptation under Continuous Knowledge Drift in the Real Worldcited by 4
  6. 2025Memorization and Knowledge Injection in Gated LLMscited by 3
  7. 2025Representation Interventions Enable Lifelong Knowledge Memory Control in LLMscited by 3
  8. 2026Higher-Order Knowledge Representations for Agentic Scientific Reasoningcited by 3
  9. 2026Scaling Laws for Hypernetwork-Based Knowledge Injection in Large Language Modelscited by 2
  10. 2025RippleBench: Capturing Ripple Effects Using Existing Knowledge Repositoriescited by 1
  11. 2026Norm Anchors Make Model Edits Lastcited by 1
  12. 2026Reversible Lifelong Model Editing via Semantic Routing-Based LoRAcited by 1
14 more
  1. 2026Modality-Decoupled Online Recursive Editingcited by 1
  2. 2025Aligning Language Models with Real-time Knowledge Editingcited by 0
  3. 2026Conflict-Resolving and Sharpness-Aware Minimization for Generalized Knowledge Editing with Multiple Updatescited by 0
  4. 2026HoReN: Normalized Hopfield Retrieval for Large-Scale Sequential Model Editingcited by 0
  5. 2026Multimodal Knowledge Edit-Scoped Generalization for Online Recursive MLLM Editingcited by 0
  6. 2026Can a Language Model Learn Facts Continually in Its Weights?cited by 0
  7. 2026KLOD: Locality-Preserving Knowledge Editing via Non-Target Distribution Preservationcited by 0
  8. 2026Synthetic Worlds for Temporal Evaluation and Knowledge Updating in LLMscited by 0
  9. 2026Popular Knowledge Propagates More Errors in LLM Knowledge Updatingcited by 0
  10. 2026The evolution of sex for artificial intelligence: a population-genetic framework for multigenerational model populationscited by 0
  11. 2026Generalizable Lifelong Model Editing via Preference Optimizationcited by 0
  12. 2026MedKIT: Evaluating Knowledge Integration and Generalization in Large Language Modelscited by 0
  13. 2026PAPER2LLM++: Continual Self-Evolution of LLMs from Research Paperscited by 0
  14. 2026LadderEdit: Edit-Level Residual Compression for Memory-Efficient Lifelong Editing of LLMscited by 0