arXiv - CS - Computation and LanguagePub Date : 2023-08-09, DOI:arxiv-2308.04645 Ercong Nie, Helmut Schmid, Hinrich Schütze 选区解析在推进自然语言处理(NLP)任务中发挥着基础作用。然而,由于为古代语言构建树库所面临的固有挑战,仅依靠带注释的解析数据来训练古代语言的自动句法分析系统是一项艰巨的任务...
arXiv - CS - Computation and Language Pub Date : 2023-08-10 , DOI: arxiv-2308.05725 Tu Anh Nguyen, Wei-Ning Hsu, Antony D'Avirro, Bowen Shi, Itai Gat, Maryam Fazel-Zarani, Tal Remez, Jade Copet, Gabriel Synnaeve, Michael Hassid, Felix Kreuk, Yossi Adi, Emmanuel Dupoux ...
Inexpensive Domain Adaptation of Pretrained Language Models: A Case Study on Biomedical Named Entity Recognition 04月08日 12:22 上传 · 自然语言 0.32MB 0 Windowing Models for Abstractive Summarization of Long Texts 04月08日 12:22 上传 · 自然语言 0.62MB 0 KorNLI and KorSTS: New Be...
We introduce two strategies to address the issue of cross-model feature corruption of existing visual prompting methods and enhance the transferability of the learned prompts, including 1) Feature Consistency Alignment: which imposes constraints to the prompted feature changes to maintain task-agnostic kn...
[1] [LMM] [Multimodal, LMM] Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models Title: Reka Core, Flash, and Edge: 一系列强大的多模态语言模型 Paper: https://arxiv.org/abs/240…
摘要:This work introduces an efficient method to scale Transformer-based Large Language Models (LLMs) to infinitely long inputs with bounded memory and computation. A key component in our proposed approach is a new attention technique dubbed Infini-attention. The Infini-attention incorporates a ...
MixSpeech: Cross-Modality Self-Learning with Audio-Visual Stream Mixup for Visual Speech Translation and Recognition arXiv - CS - Computation and LanguagePub Date : 2023-03-09, DOI:arxiv-2303.05309 Xize Cheng, Linjun Li, Tao Jin, Rongjie Huang, Wang Lin, Zehan Wang, Huangdai Liu, Ye Wan...
从arXiv搬运的2020年03月26日的Computation and Language领域论文合集打包18篇 论文 | 0人 订阅 | 0人 下载 订阅 The Medical Scribe: Corpus Development and Model Performance Analyses 03月26日 12:27 上传 · 自然语言 0.77MB 1 Meta-CoTGAN: A Meta Cooperative Training Paradigm for Improving ...
arXiv - CS - Computation and Language Pub Date : 2021-09-22 , DOI: arxiv-2109.10855 Zichao Li, Dheeraj Mekala, Chengyu Dong, Jingbo Shang 后门攻击通过注入触发器和修改标签使训练数据的子集中毒,从而将人为漏洞引入模型中。已经探索了各种触发器设计策略来攻击文本分类器,但是,防御此类攻击仍然是一个悬...
随着移动计算和网络技术的快速发展,攻击性语言在社交网络平台上变得更加普遍。由于本地语言中的攻击性语言识别对于调节社交媒体内容至关重要,因此在本文中,我们使用了三种资源不足的达罗毗荼语言,即马拉雅拉姆语、泰米尔语和卡纳达语。我们在 FIRE 2020-HASOC-DravidianCodeMix 和 EACL 2021 的 DravidianLangTech 上提出了...