paper-with-me

Papers

Deep Bidirectional Language-Knowledge Graph Pretraining

2022-10-17 · Michihiro Yasunaga, Antoine Bosselut, Hongyu Ren, Xikun Zhang, Christopher D Manning, Percy Liang, Jure Leskovec

Pretraining a language model (LM) on text has been shown to help various downstream NLP tasks. Recent works show that a knowledge graph (KG) can complement text data, offering structured background knowledge that provides a useful scaffold for reasoning. However, these works are not pretrained to learn a deep fusion of the two modalities at scale, limiting the potential to acquire fully joint representations of text and KG. Here we propose DRAGON (Deep Bidirectional Language-Knowledge Graph Pretraining), a self-supervised approach to pretraining a deeply joint language-knowledge foundation model from text and KG at scale. Specifically, our model takes pairs of text segments and relevant KG subgraphs as input and bidirectionally fuses information from both modalities. We pretrain this model by unifying two self-supervised reasoning tasks, masked language modeling and KG link prediction. DRAGON outperforms existing LM and LM+KG models on diverse downstream tasks including question answering across general and biomedical domains, with +5% absolute gain on average. In particular, DRAGON achieves notable performance on complex reasoning about language and knowledge (+10% on questions involving long contexts or multi-step reasoning) and low-resource QA (+8% on OBQA and RiddleSense), and new state-of-the-art results on various BioNLP tasks. Our code and trained models are available at https://github.com/michiyasunaga/dragon.

📄 PDF Abstract BibTeX arXiv:2210.09338

Code (2)

michiyasunaga/dragon 공식 구현 pytorch
HaochenLiu2000/QAP pytorch

Tasks

Common Sense ReasoningKnowledge GraphsLanguage ModelingLanguage ModellingLink PredictionMasked Language ModelingQuestion AnsweringRiddle Sense

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Residual Connection 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
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…
Adam 설명 없음

Similar Papers 제목 키워드 기반

XLNet: Generalized Autoregressive Pretraining for Language Understanding

2019-06-19 · NeurIPS 2019 12 · Zhilin Yang, Zihang Dai, Yiming Yang, Jaime Carbonell 외

With the capability of modeling bidirectional contexts, denoising autoencoding based pretraining like BERT achieves better performance than pretraining approaches based on autoregressive language modeling. However, relyi…

Audio Question AnsweringChinese Reading ComprehensionDocument RankingHumor Detection+9

Data-Efficient Autoregressive-to-Diffusion Language Models via On-Policy Distillation

2026-06-04 · Xingyu Su, Jacob Helwig, Shubham Parashar, Atharv Chagi 외 arxiv

We study the transformation of autoregressive models (ARLMs) into diffusion language models (DLMs). Rather than pretraining from scratch, prior work replaces the causal attention in ARLMs with bidirectional attention and…

ALBERT with Knowledge Graph Encoder Utilizing Semantic Similarity for Commonsense Question Answering

2022-11-14 · Byeongmin Choi, YongHyun Lee, Yeunwoong Kyung, Eunchan Kim

Recently, pre-trained language representation models such as bidirectional encoder representations from transformers (BERT) have been performing well in commonsense question answering (CSQA). However, there is a problem …

Language ModelingLanguage ModellingQuestion AnsweringRelation Network+2

Syntactic Structure Distillation Pretraining For Bidirectional Encoders

2020-05-27 · Adhiguna Kuncoro, Lingpeng Kong, Daniel Fried, Dani Yogatama 외

Textual representation learners trained on large amounts of data have achieved notable success on downstream tasks; intriguingly, they have also performed well on challenging tests of syntactic competence. Given this suc…

Knowledge DistillationLanguage ModelingLanguage ModellingNatural Language Understanding+2

Improving Medical Visual Representations via Radiology Report Generation

2023-10-30 · Keegan Quigley, Miriam Cha, Josh Barua, Geeticka Chauhan 외

Vision-language pretraining has been shown to produce high-quality visual encoders which transfer efficiently to downstream computer vision tasks. Contrastive learning approaches have increasingly been adopted for medica…

Contrastive LearningDecoderImage CaptioningMedical Image Analysis