paper-with-me

Papers

NextLevelBERT: Masked Language Modeling with Higher-Level Representations for Long Documents

2024-02-27 · Tamara Czinczoll, Christoph Hönes, Maximilian Schall, Gerard de Melo

While (large) language models have significantly improved over the last years, they still struggle to sensibly process long sequences found, e.g., in books, due to the quadratic scaling of the underlying attention mechanism. To address this, we propose NextLevelBERT, a Masked Language Model operating not on tokens, but on higher-level semantic representations in the form of text embeddings. We pretrain NextLevelBERT to predict the vector representation of entire masked text chunks and evaluate the effectiveness of the resulting document vectors on three types of tasks: 1) Semantic Textual Similarity via zero-shot document embeddings, 2) Long document classification, 3) Multiple-choice question answering. We find that next-level Masked Language Modeling is an effective technique to tackle long-document use cases and can outperfor much larger embedding models as long as the required level of detail of semantic information is not too fine. Our models and code are publicly available online.

📄 PDF Abstract BibTeX arXiv:2402.17682

Code (1)

aiintelligentsystems/next-level-bert 공식 구현 pytorch

Tasks

Document ClassificationLanguage ModelingLanguage ModellingMasked Language ModelingMultiple-choiceQuestion AnsweringSemantic Textual Similarity

Similar Papers 제목 키워드 기반

MeLT: Message-Level Transformer with Masked Document Representations as Pre-Training for Stance Detection

2021-09-16 · Findings (EMNLP) 2021 11 · Matthew Matero, Nikita Soni, Niranjan Balasubramanian, H. Andrew Schwartz

Much of natural language processing is focused on leveraging large capacity language models, typically trained over single messages with a task of predicting one or more tokens. However, modeling human language at higher…

AttributeLanguage ModelingLanguage ModellingMasked Language Modeling+1

LightCLIP: Learning Multi-Level Interaction for Lightweight Vision-Language Models

2023-12-01 · Ying Nie, wei he, Kai Han, Yehui Tang 외

Vision-language pre-training like CLIP has shown promising performance on various downstream tasks such as zero-shot image classification and image-text retrieval. Most of the existing CLIP-alike works usually adopt rela…

image-classificationImage ClassificationImage-text RetrievalLanguage Modeling+4

MAMO: Masked Multimodal Modeling for Fine-Grained Vision-Language Representation Learning

2022-10-09 · Zijia Zhao, Longteng Guo, Xingjian He, Shuai Shao 외

Multimodal representation learning has shown promising improvements on various vision-language tasks. Most existing methods excel at building global-level alignment between vision and language while lacking effective fin…

Image-text Retrievalmultimodal interactionQuestion AnsweringRepresentation Learning+6

HybridMIM: A Hybrid Masked Image Modeling Framework for 3D Medical Image Segmentation

2023-03-18 · Zhaohu Xing, Lei Zhu, Lequan Yu, Zhiheng Xing 외

Masked image modeling (MIM) with transformer backbones has recently been exploited as a powerful self-supervised pre-training technique. The existing MIM methods adopt the strategy to mask random patches of the image and…

Contrastive LearningImage SegmentationMedical Image SegmentationSelf-Supervised Learning+1

MaDiS: Taming Masked Diffusion Language Models for Sign Language Generation

2026-01-27 · Ronglai Zuo, Rolandos Alexandros Potamias, Qi Sun, Evangelos Ververas 외 arxiv

Sign language generation (SLG) aims to translate written texts into expressive sign motions, bridging communication barriers for the Deaf and Hard-of-Hearing communities. Recent studies formulate SLG within the language …