arXiv AI
Golden Layers and Where to Find Them: Improved Knowledge Editing for Large Language Models Via Layer Gradient Analysis
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Jun 19
LOKI: Memory-Free Null-Space Constrained Lifelong Knowledge Editing
arXiv:2606. 19679v1 Announce Type: cross Abstract: Lifelong knowledge editing aims to efficiently and sequentially update language models over time, as new knowledge becomes available or when the model makes mistakes, while preserving acceptable performance on past knowledge.
arXiv AI
Oct 26
MetaKE: Meta-Learning for Knowledge Editing Toward a Better Accuracy-Editability Trade-off
arXiv:2603.12677v4 Announce Type: replace-cross Abstract: Existing locate-then-edit Knowledge Editing (KE) methods typically decompose editing into two stages: upstream target representation optimiza...
arXiv AI
Jun 2
Revisiting Parameter-Based Knowledge Editing in Large Language Models: Theoretical Limits and Empirical Evidence
arXiv:2606. 00570v1 Announce Type: cross Abstract: Parameter-based knowledge editing updates the internal knowledge of large language models (LLMs) via localized weight modifications and has attracted significant attention.
arXiv AI
Oct 27
DynEdit: Dynamic Entropy-Guided Sequential Editing for Large Language Models
DynEdit is a dynamic framework for editing long-sequence knowledge in large language models. It uses token-level entropy to guide non-uniform sliding windows, focusing on high-entropy, information-dense regions that are harder to edit. Experiments on three benchmarks and two backbone models show DynEdit outperforms baselines, achieving up to +14.0 BLEU points on challenging paraphrase queries and near-saturated semantic scores across diverse knowledge domains.
arXiv AI
Jun 10
Benchmarking Knowledge Editing using Logical Rules
arXiv:2606. 10554v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly deployed in real-world applications that require access to up-to-date knowledge.
arXiv AI
Oct 5
Improving Atomic-Fact Recall via Focused Views in Unstructured Knowledge Editing
The paper introduces FOVEATED, a plug‑and‑play framework that improves atomic‑fact recall in unstructured knowledge editing (UKE) for large language models. By randomly shifting Rotary Position Embedding (RoPE) positions during editing, FOVEATED creates focused views of each sentence, counteracting the context‑reliance problem where edited LLMs reproduce passages but fail to recall individual facts. Experiments show consistent gains across five editors, two LLM backbones, and three benchmarks.