paper-with-me

Papers

Knowledge Neurons in Pretrained Transformers

2021-04-18 · ACL 2022 5 · Damai Dai, Li Dong, Yaru Hao, Zhifang Sui, Baobao Chang, Furu Wei

Large-scale pretrained language models are surprisingly good at recalling factual knowledge presented in the training corpus. In this paper, we present preliminary studies on how factual knowledge is stored in pretrained Transformers by introducing the concept of knowledge neurons. Specifically, we examine the fill-in-the-blank cloze task for BERT. Given a relational fact, we propose a knowledge attribution method to identify the neurons that express the fact. We find that the activation of such knowledge neurons is positively correlated to the expression of their corresponding facts. In our case studies, we attempt to leverage knowledge neurons to edit (such as update, and erase) specific factual knowledge without fine-tuning. Our results shed light on understanding the storage of knowledge within pretrained Transformers. The code is available at https://github.com/Hunter-DDM/knowledge-neurons.

📄 PDF Abstract BibTeX arXiv:2104.08696

Code (3)

hunter-ddm/knowledge-neurons 공식 구현 pytorch
EleutherAI/knowledge-neurons pytorch
kmeng01/rome pytorch

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Weight Decay 설명 없음
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Residual Connection 설명 없음
Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.

Similar Papers 제목 키워드 기반

Knowledge Neurons in Pretrained Transformers

2021-09-17 · ACL ARR September 2021 9 · Anonymous

Large-scale pretrained language models are surprisingly good at recalling factual knowledge presented in the training corpus. In this paper, we present preliminary studies on how factual knowledge is stored in pretrained…

Identifying Multi-modal Knowledge Neurons in Pretrained Transformers via Two-stage Filtering

2025-03-29 · Yugen Sato, Tomohiro Takagi

Recent advances in large language models (LLMs) have led to the development of multimodal LLMs (MLLMs) in the fields of natural language processing (NLP) and computer vision. Although these models allow for integrated vi…

Caption Generationknowledge editingMisinformation

DEPN: Detecting and Editing Privacy Neurons in Pretrained Language Models

2023-10-31 · Xinwei Wu, Junzhuo Li, Minghui Xu, Weilong Dong 외

Large language models pretrained on a huge amount of data capture rich knowledge and information in the training data. The ability of data memorization and regurgitation in pretrained language models, revealed in previou…

MemorizationModel Editing

Tracing Stereotypes in Pre-trained Transformers: From Biased Neurons to Fairer Models

2026-01-09 · Gianmario Voria, Moses Openja, Foutse Khomh, Gemma Catolino 외 arxiv

The advent of transformer-based language models has reshaped how AI systems process and generate text. In software engineering (SE), these models now support diverse activities, accelerating automation and decision-makin…

Neural Knowledge Bank for Pretrained Transformers

2022-07-31 · Damai Dai, Wenbin Jiang, Qingxiu Dong, Yajuan Lyu 외

The ability of pretrained Transformers to remember factual knowledge is essential but still limited for existing models. Inspired by existing work that regards Feed-Forward Networks (FFNs) in Transformers as key-value me…

Language ModelingLanguage ModellingMachine TranslationQuestion Answering+1