paper-with-me

홈 › Papers

CaKE: Circuit-aware Editing Enables Generalizable Knowledge Learners

2025-03-20 · Yunzhi Yao, Jizhan Fang, Jia-Chen Gu, Ningyu Zhang, Shumin Deng, Huajun Chen, Nanyun Peng

Knowledge Editing (KE) enables the modification of outdated or incorrect information in large language models (LLMs). While existing KE methods can update isolated facts, they struggle to generalize these updates to multi-hop reasoning tasks that depend on the modified knowledge. Through an analysis of reasoning circuits -- the neural pathways LLMs use for knowledge-based inference, we observe that current layer-localized KE approaches, such as MEMIT and WISE, which edit only single or a few model layers, struggle to effectively incorporate updated information into these reasoning pathways. To address this limitation, we propose CaKE (Circuit-aware Knowledge Editing), a novel method that enables more effective integration of updated knowledge in LLMs. CaKE leverages strategically curated data, guided by our circuits-based analysis, that enforces the model to utilize the modified knowledge, stimulating the model to develop appropriate reasoning circuits for newly integrated knowledge. Experimental results show that CaKE enables more accurate and consistent use of updated knowledge across related reasoning tasks, leading to an average of 20% improvement in multi-hop reasoning accuracy on MQuAKE dataset compared to existing KE methods. We release the code and data in https://github.com/zjunlp/CaKE.

📄 PDF Abstract BibTeX arXiv:2503.16356

Code (1)

zjunlp/cake 공식 구현 pytorch

Tasks

knowledge editing

Similar Papers 제목 키워드 기반

SCAN: Sparse Circuit Anchor Interpretable Neuron for Lifelong Knowledge Editing

2026-03-16 · Yuhuan Liu, Haitian Zhong, Xinyuan Xia, Qiang Liu 외 arxiv

Large Language Models (LLMs) often suffer from catastrophic forgetting and collapse during sequential knowledge editing. This vulnerability stems from the prevailing dense editing paradigm, which treats models as black b…

knowledge editing

Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving

2024-06-24 · Ruoyu Qin, Zheming Li, Weiran He, Mingxing Zhang 외

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI. It features a KVCache-centric disaggregated architecture that separates the prefill and decoding clusters. It also leverages the u…

CPUGPU

3D GAN Inversion with Pose Optimization

2022-10-13 · Jaehoon Ko, Kyusun Cho, Daewon Choi, Kwangrok Ryoo 외

With the recent advances in NeRF-based 3D aware GANs quality, projecting an image into the latent space of these 3D-aware GANs has a natural advantage over 2D GAN inversion: not only does it allow multi-view consistent e…

3D ReconstructionImage ReconstructionNeRFNovel View Synthesis

Addressing the Reasoning Gap: Mechanistic Circuit-Based Knowledge Editing in Large Language Models

2026-04-07 · Tianyi Zhao, Yinhan He, Wendy Zheng, Chen Chen arxiv

Deploying Large Language Models (LLMs) in real-world dynamic environments raises the challenge of updating their pre-trained knowledge. While existing knowledge editing methods can reliably patch isolated facts, they fre…

knowledge editing

LightCAKE: A Lightweight Framework for Context-Aware Knowledge Graph Embedding

2021-02-22 · Zhiyuan Ning, Ziyue Qiao, Hao Dong, Yi Du 외

Knowledge graph embedding (KGE) models learn to project symbolic entities and relations into a continuous vector space based on the observed triplets. However, existing KGE models cannot make a proper trade-off between t…

Graph EmbeddingKnowledge Graph EmbeddingKnowledge Graphs