paper-with-me

Papers

Knowledge-Aware Language Model Pretraining

2020-06-29 · Corby Rosset, Chenyan Xiong, Minh Phan, Xia Song, Paul Bennett, Saurabh Tiwary

How much knowledge do pretrained language models hold? Recent research observed that pretrained transformers are adept at modeling semantics but it is unclear to what degree they grasp human knowledge, or how to ensure they do so. In this paper we incorporate knowledge-awareness in language model pretraining without changing the transformer architecture, inserting explicit knowledge layers, or adding external storage of semantic information. Rather, we simply signal the existence of entities to the input of the transformer in pretraining, with an entity-extended tokenizer; and at the output, with an additional entity prediction task. Our experiments show that solely by adding these entity signals in pretraining, significantly more knowledge is packed into the transformer parameters: we observe improved language modeling accuracy, factual correctness in LAMA knowledge probing tasks, and semantics in the hidden representations through edge probing.We also show that our knowledge-aware language model (KALM) can serve as a drop-in replacement for GPT-2 models, significantly improving downstream tasks like zero-shot question-answering with no task-related training.

📄 PDF Abstract BibTeX arXiv:2007.00655

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge ProbingLanguage ModelingLanguage ModellingmodelQuestion Answering

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…
Discriminative Fine-Tuning Discriminative Fine-Tuning is a fine-tuning strategy that is used for ULMFiT type models. Instead of using the same learning rate…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Multi-Head Attention 설명 없음
Residual Connection 설명 없음
Attention 설명 없음

Similar Papers 제목 키워드 기반

Source-Aware Training Enables Knowledge Attribution in Language Models

2024-04-01 · Muhammad Khalifa, David Wadden, Emma Strubell, Honglak Lee 외

Large language models (LLMs) learn a vast amount of knowledge during pretraining, but they are often oblivious to the source(s) of such knowledge. We investigate the problem of intrinsic source citation, where LLMs are r…

Data Augmentation

Pretrain Knowledge-Aware Language Models

2021-01-01 · Corbin L Rosset, Chenyan Xiong, Minh Phan, Xia Song 외

How much knowledge do pretrained language models hold? Recent research observed that pretrained transformers are adept at modeling semantics but it is unclear to what degree they grasp human knowledge, or how to ensure t…

Knowledge ProbingLanguage ModelingLanguage ModellingQuestion Answering

Sem4SAP: Synonymous Expression Mining From Open Knowledge Graph For Language Model Synonym-Aware Pretraining

2023-03-25 · Zhouhong Gu, Sihang Jiang, Wenhao Huang, Jiaqing Liang 외

The model's ability to understand synonymous expression is crucial in many kinds of downstream tasks. It will make the model to better understand the similarity between context, and more robust to the synonym substitutio…

Language ModelingLanguage Modelling

Parametric Knowledge is Not All You Need: Toward Honest Large Language Models via Retrieval of Pretraining Data

2026-01-29 · Christopher Adrian Kusuma, Muhammad Reza Qorib, Hwee Tou Ng arxiv

Large language models (LLMs) are highly capable of answering questions, but they are often unaware of their own knowledge boundary, i.e., knowing what they know and what they don't know. As a result, they can generate fa…

Meta-learning for downstream aware and agnostic pretraining

2021-06-06 · Hongyin Luo, Shuyan Dong, Yung-Sung Chuang, Shang-Wen Li

Neural network pretraining is gaining attention due to its outstanding performance in natural language processing applications. However, pretraining usually leverages predefined task sequences to learn general linguistic…

Meta-Learning