REACT: Representation Extraction And Controllable Tuning to Overcome Overfitting in LLM Knowledge Editing

• URL: http://arxiv.org/abs/2505.18933v1
• Date: Sun, 25 May 2025 01:57:06 GMT
• Title: REACT: Representation Extraction And Controllable Tuning to Overcome Overfitting in LLM Knowledge Editing
• Abstract summary: We introduce REACT, a framework for precise and controllable knowledge editing.<n>In the initial phase, we utilize tailored stimuli to extract latent factual representations.<n>In the second phase, we apply controllable perturbations to hidden states using the obtained vector with a magnitude scalar.
• Score: 42.89229070245538
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language model editing methods frequently suffer from overfitting, wherein factual updates can propagate beyond their intended scope, overemphasizing the edited target even when it's contextually inappropriate. To address this challenge, we introduce REACT (Representation Extraction And Controllable Tuning), a unified two-phase framework designed for precise and controllable knowledge editing. In the initial phase, we utilize tailored stimuli to extract latent factual representations and apply Principal Component Analysis with a simple learnbale linear transformation to compute a directional "belief shift" vector for each instance. In the second phase, we apply controllable perturbations to hidden states using the obtained vector with a magnitude scalar, gated by a pre-trained classifier that permits edits only when contextually necessary. Relevant experiments on EVOKE benchmarks demonstrate that REACT significantly reduces overfitting across nearly all evaluation metrics, and experiments on COUNTERFACT and MQuAKE shows that our method preserves balanced basic editing performance (reliability, locality, and generality) under diverse editing scenarios.

Related papers

• Orthogonal Representation Editing: Decoupling Semantic Entanglement in Batch Knowledge Editing of LLMs [38.268937785139855]
  Knowledge editing aims to efficiently update factual information in Large Language Models (LLMs) without full retraining.<n>Existing methods still suffer from performance degradation in batch knowledge editing.<n>We propose Orthogonal Representation Editing (ORE), which performs edits in the hidden representation space of LLMs.
  arXiv  Detail & Related papers  (2026-06-21T18:29:40Z)
• EditSR: Enhancing Neural Symbolic Regression via Edit-based Rectification [11.013304470971017]
  We propose EditSR, a two-layer framework that combines a neural symbolic regression model in the first layer with an edit-based Rectifier in the second layer to achieve efficient prediction and post-hoc rectification.<n>To ensure syntactic validity throughout rectification, each edit action is restricted to a syntactically valid space so that every edited expression remains parseable.
  arXiv  Detail & Related papers  (2026-06-06T00:37:03Z)
• The Anatomy of an Edit: Mechanism-Guided Activation Steering for Knowledge Editing [52.315369633116255]
  Large language models (LLMs) are increasingly used as knowledge bases, but keeping them up to date requires targeted knowledge editing (KE)<n>We take a mechanistic view of KE using neuron-level knowledge attribution (NLKA)<n>Across representative KE methods, we find a consistent pattern: mid-to-late attention predominantly promotes the new target, while attention and FFN modules cooperate to suppress the original fact.<n>Motivated by these findings, we propose MEGA, a MEchanism-Guided Activation steering method that performs attention-residual interventions in attribution-aligned regions.
  arXiv  Detail & Related papers  (2026-03-21T12:40:15Z)
• MetaKE: Meta-learning Aligned Knowledge Editing via Bi-level Optimization [11.430206422495829]
  State-of-the-art methods suffer from an open-loop control mismatch.<n>We identify a critical "Semantic-Execution Disconnect"<n>We propose MetaKE, a new framework that reframes KE as a bi-level optimization problem.
  arXiv  Detail & Related papers  (2026-03-13T05:47:00Z)
• Generalizable Multimodal Large Language Model Editing via Invariant Trajectory Learning [46.514554089834554]
  Existing editing methods rely on a rigid mapping from parameter or module modifications to output.<n>In this paper, we reformulate MLLM editing as an out-of-distribution (OOD) generalization problem.<n>We propose ODEdit, a plug-and-play invariant learning based framework that enhances editing reliability, locality, and generality.
  arXiv  Detail & Related papers  (2026-01-27T15:25:07Z)
• EtCon: Edit-then-Consolidate for Reliable Knowledge Editing [85.20993502078899]
  We propose Edit-then-Consolidate, a novel knowledge editing paradigm that aims to bridge the gap between theoretical knowledge editing methods and their real-world applicability.<n>Our framework consistently improves editing reliability and generalization under real-world evaluations, while better preserving locality and pre-trained capabilities.
  arXiv  Detail & Related papers  (2025-12-04T12:43:50Z)
• Representation Interventions Enable Lifelong Unstructured Knowledge Control [54.86207134539453]
  Large language models (LLMs) often produce incorrect or outdated content. Updating their knowledge efficiently and accurately without costly retraining is a major challenge.<n>We introduce RILKE, a robust and scalable method that treats knowledge control as interventions within the model's representation space.<n>During training, RILKE learns paraphrase-robust and edit-localized modules that limit each update to a low-dimensional subspace to minimize cross-edit interference.<n>In inference, a query-adaptive router selects the appropriate module to guide the model's generation.
  arXiv  Detail & Related papers  (2025-11-25T22:15:00Z)
• MEMOIR: Lifelong Model Editing with Minimal Overwrite and Informed Retention for LLMs [82.34547399693966]
  Existing methods for lifelong model editing compromise generalization, interfere with past edits, or fail to scale to long editing sequences.<n>We propose MEMOIR, a novel scalable framework that injects knowledge through a residual memory.<n>MeMOIR confines each edit to a distinct subset of the memory parameters, minimizing interference among edits.
  arXiv  Detail & Related papers  (2025-06-09T16:16:42Z)
• Tracing and Reversing Rank-One Model Edits [5.260519479124422]
  This work investigates the traceability and reversibility of knowledge edits, focusing on the widely used Rank-One Model Editing (ROME) method.<n>We show that ROME introduces distinctive distributional patterns in the edited weight matrices, which can serve as effective signals for locating the edited weights.<n>We propose a method to infer the edited object entity directly from the modified weights, without access to the editing prompt, achieving over 95% accuracy.
  arXiv  Detail & Related papers  (2025-05-27T07:27:01Z)
• Learning Where to Edit Vision Transformers [27.038720045544867]
  We propose a locate-then-edit approach for editing vision Transformers (ViTs) in computer vision. We first address the where-to-edit challenge by meta-learning a hypernetwork on CutMix-augmented data. To validate our method, we construct an editing benchmark that introduces subpopulation shifts towards natural underrepresented images and AI-generated images.
  arXiv  Detail & Related papers  (2024-11-04T10:17:40Z)
• Uncovering Overfitting in Large Language Model Editing [35.55260822503773]
  We identify and investigate the phenomenon of Editing Overfit, where edited models assign disproportionately high probabilities to the edit target. We propose a new plug-and-play strategy called Learn to Inference (LTI), which introduce a Multi-stage Inference Constraint module to guide the edited models in recalling new knowledge.
  arXiv  Detail & Related papers  (2024-10-10T11:09:00Z)
• PseudoNeg-MAE: Self-Supervised Point Cloud Learning using Conditional Pseudo-Negative Embeddings [55.55445978692678]
  PseudoNeg-MAE enhances global feature representation of point cloud masked autoencoders by making them both discriminative and sensitive to transformations.<n>We propose a novel loss that explicitly penalizes invariant collapse, enabling the network to capture richer transformation cues while preserving discriminative representations.
  arXiv  Detail & Related papers  (2024-09-24T07:57:21Z)
• The Butterfly Effect of Model Editing: Few Edits Can Trigger Large Language Models Collapse [58.0132400208411]
  Even a single edit can trigger model collapse, manifesting as significant performance degradation in various benchmark tasks. benchmarking Large Language Models after each edit is impractically time-consuming and resource-intensive. We have utilized GPT-3.5 to develop a new dataset, HardEdit, based on hard cases.
  arXiv  Detail & Related papers  (2024-02-15T01:50:38Z)
• SWEA: Updating Factual Knowledge in Large Language Models via Subject Word Embedding Altering [17.20346072074533]
  Recent model editing is a promising technique for efficiently updating a small amount of knowledge of large language models.<n>We propose a detachable and expandable Subject Word Embedding Altering (SWEA) framework, which finds the editing embeddings through token-level matching.<n>We demonstrate the overall state-of-the-art (SOTA) performance of SWEA$oplus$OS on the CounterFact and zsRE datasets.
  arXiv  Detail & Related papers  (2024-01-31T13:08:45Z)
• Model Editing Harms General Abilities of Large Language Models: Regularization to the Rescue [122.20016030723043]
  We evaluate the side effects of model editing on large language models (LLMs) Our analysis reveals that the side effects are caused by model editing altering the original model weights excessively. To mitigate this, a method named RECT is proposed to regularize the edit update weights.
  arXiv  Detail & Related papers  (2024-01-09T18:03:15Z)
• Test-Time Training for Semantic Segmentation with Output Contrastive Loss [12.535720010867538]
  Deep learning-based segmentation models have achieved impressive performance on public benchmarks, but generalizing well to unseen environments remains a major challenge. This paper introduces Contrastive Loss (OCL), known for its capability to learn robust and generalized representations, to stabilize the adaptation process. Our method excels even when applied to models initially pre-trained using domain adaptation methods on test domain data, showcasing its resilience and adaptability.
  arXiv  Detail & Related papers  (2023-11-14T03:13:47Z)
• On the Limits of Evaluating Embodied Agent Model Generalization Using Validation Sets [101.28658250723804]
  This paper experiments with augmenting a transformer model with modules that effectively utilize a wider field of view and learn to choose whether the next step requires a navigation or manipulation action. We observe that the proposed modules resulted in improved, and in fact state-of-the-art performance on an unseen validation set of a popular benchmark dataset, ALFRED. We highlight this result as we believe it may be a wider phenomenon in machine learning tasks but primarily noticeable only in benchmarks that limit evaluations on test splits.
  arXiv  Detail & Related papers  (2022-05-18T23:52:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.