知乎博客检索

具有交叉注意力控制功能的提示-提示图像编辑
https://zhuanlan.zhihu.com/p/577527077
PROMPT-TO-PROMPT IMAGE EDITING WITH CROSS-ATTENTION CONTROL

知识图谱在食品科学与工业中的应用
https://zhuanlan.zhihu.com/p/591966904
APPLICATIONS OF KNOWLEDGE GRAPHS FOR THE FOOD SCIENCE AND INDUSTRY

AltCLIP：改变CLIP中的语言编码器以扩展语言能力
https://zhuanlan.zhihu.com/p/589700046
AltCLIP：Altering the Language Encoder in CLIP for Extended Language Capabilities

使用Attention-RPN和多关系检测器进行few-shot目标检测
https://zhuanlan.zhihu.com/p/588230585
Few-Shot Object Detection with Attention-RPN and Multi-Relation Detector

FSCE: 通过对比性建议编码进行的few-shot目标检测
https://zhuanlan.zhihu.com/p/588231035
FSCE： Few-Shot Object Detection via Contrastive Proposal Encoding

AliMe Assist：打造创新电商体验的智能助手
https://zhuanlan.zhihu.com/p/587596523
AliMe Assist： An Intelligent Assistant for Creating an Innovative E-commerce Experience

社交聊天机器人小冰的设计与实现
https://zhuanlan.zhihu.com/p/587595835
The Design and Implementation of XiaoIce, an Empathetic Social Chatbot

用于多目标跟踪的准密集相似性学习
https://zhuanlan.zhihu.com/p/586082025
Quasi-Dense Similarity Learning for Multiple Object Tracking

深度感知的生成对抗网络用于口播视频的生成
https://zhuanlan.zhihu.com/p/569320116
Depth-Aware Generative Adversarial Network for Talking Head Video Generation

使用 Vision Transformers 进行简单的开放式单词表目标检测
https://zhuanlan.zhihu.com/p/586087658
Simple Open-Vocabulary Object Detection with Vision Transformers

用于one-shot目标检测的平衡和层次关系学习
https://zhuanlan.zhihu.com/p/586094468
Balanced and Hierarchical Relation Learning for One-shot Object Detection

用于one-shot目标检测的自适应图像transformer
https://zhuanlan.zhihu.com/p/585327946
Adaptive Image Transformer for One-Shot Object Detection

roformer：带有旋转位置嵌入的增强型transformer
https://zhuanlan.zhihu.com/p/574478161
ROFORMER： ENHANCED TRANSFORMER WITH ROTARY POSITION EMBEDDING

用于One-shot目标检测的语义对齐融合transformer
https://zhuanlan.zhihu.com/p/585256263
Semantic-aligned Fusion Transformer for One-shot Object Detection

PIFuHD：用于高分辨率三维人体数字化的多级像素对齐隐式函数
https://zhuanlan.zhihu.com/p/566093991
PIFuHD： Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization

Imagic: 基于文本的真实图像编辑与扩散模型
https://zhuanlan.zhihu.com/p/576710237
Imagic: Text-Based Real Image Editing with Diffusion Models

多模态对比学习LIMoE: 图像-语言的混合专家
https://zhuanlan.zhihu.com/p/583728857
Multimodal Contrastive Learning with LIMoE：the Language-Image Mixture of Experts

统一的多选视角实现自然语言理解的zero-shot学习
https://zhuanlan.zhihu.com/p/577097077
Zero-Shot Learners for Natural Language Understanding via a Unified Multiple Choice Perspective

用于通用信息提取的统一结构生成
https://zhuanlan.zhihu.com/p/569268582
Unified Structure Generation for Universal Information Extraction

Copilot: 评估在代码上训练的大型语言模型
https://zhuanlan.zhihu.com/p/571373422
Evaluating Large Language Models Trained on Code

视觉表示学习的多模态对比性训练
https://zhuanlan.zhihu.com/p/544355035
Multimodal Contrastive Training for Visual Representation Learning

KenLM：更快、更小的语言模型查询
https://zhuanlan.zhihu.com/p/564731709
KenLM： Faster and Smaller Language Model Queries

GenIE: 生成式信息提取
https://zhuanlan.zhihu.com/p/562155662
GenIE： Generative Information Extraction

通过路由不确定性意识的交易专家进行量化股票投资的多任务学习方法
https://zhuanlan.zhihu.com/p/543453690
Quantitative Stock Investment by Routing Uncertainty-Aware Trading Experts： A Multi-Task Learning Approach

ViLBERT: 视觉和语言任务的预训练任务无关的视觉语言学表示
https://zhuanlan.zhihu.com/p/545869261
ViLBERT： Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks

用于手写2202数学表达式识别的计数感知网络
https://zhuanlan.zhihu.com/p/546590327
When Counting Meets HMER：Counting-Aware Network for Handwritten 2202 Mathematical Expression Recognition

OpenPose: 使用部分亲和域的实时多人二维姿势估计
https://zhuanlan.zhihu.com/p/561804021
OpenPose： Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields

使用二维动漫人物表的协作式神经渲染模型
https://zhuanlan.zhihu.com/p/560585021
Collaborative Neural Rendering using 2D Anime Character Sheets

GLM: 自回归空白填充的通用语言模型预训练
https://zhuanlan.zhihu.com/p/560559133
GLM： General Language Model Pretraining with Autoregressive Blank Infilling

图像作为一种外语: 所有视觉和视觉语言任务的BEIT预训练
https://zhuanlan.zhihu.com/p/559116135
Image as a Foreign Language： BEIT Pretraining for All Vision and Vision-Language Tasks

多粒度视觉语言预训练：将文本与视觉概念联系起来
https://zhuanlan.zhihu.com/p/554130166
Multi-Grained Vision Language Pre-Training：Aligning Texts with Visual Concepts

ERNIE-ViL：通过场景图的知识强化视觉语言表述
https://zhuanlan.zhihu.com/p/554100902
ERNIE-ViL： Knowledge Enhanced Vision-Language Representations through Scene Graphs

OFA：通过一个简单的seq2seq的学习框架来统一架构、任务和模态
https://zhuanlan.zhihu.com/p/548392602
OFA： UNIFYING ARCHITECTURES, TASKS, AND MODALITIES THROUGH A SIMPLE SEQUENCE-TO-SEQUENCE LEARNING FRAMEWORK

Wukong：一亿规模的中文跨模态预训练基准
https://zhuanlan.zhihu.com/p/551622338
Wukong：A 100 Million Large-scale Chinese Cross-modal Pre-training Benchmark

YOLOv7：可训练的bag-of-freebies为实时目标检测器树立了新的榜样
https://zhuanlan.zhihu.com/p/546609857
YOLOv7： Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

ReFNet:多模态融合精炼网络
https://zhuanlan.zhihu.com/p/545135269
Multimodal Fusion Refiner Networks

SimCLR:视觉表示对比学习的简单框架
https://zhuanlan.zhihu.com/p/544005001
A Simple Framework for Contrastive Learning of Visual Representations

有监督的对比学习
https://zhuanlan.zhihu.com/p/543961298
Supervised Contrastive Learning

仅从字幕中训练视觉语言transformer模型
https://zhuanlan.zhihu.com/p/540718732
Training Vision-Language Transformers from Captions Alone

MS-COCO的扩展模态内和模态间语义相似性判断
https://zhuanlan.zhihu.com/p/540954741
Crisscrossed Captions：Extended Intramodal and Intermodal Semantic Similarity Judgments for MS-COCO

利用表示编码簿进行多模态对齐
https://zhuanlan.zhihu.com/p/540703003
Multi-modal Alignment using Representation Codebook

M6: 一个中文的多模态预训练模型
https://zhuanlan.zhihu.com/p/541143888
M6： A Chinese Multimodal Pretrainer

通过文本生成将视觉和语言任务统一起来
https://zhuanlan.zhihu.com/p/540679125
Unifying Vision-and-Language Tasks via Text Generation

利用噪声文本监督扩大视觉和视觉语言表示学习的规模
https://zhuanlan.zhihu.com/p/540570838
Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision

Pixel-BERT:通过深度多模态变换将图像像素与文本对齐
https://zhuanlan.zhihu.com/p/539102716
Pixel-BERT: Aligning Image Pixels with Text by Deep Multi-Modal Transformers

通过模型不确定性融合的多任务密集检索用于开放领域的问答
https://zhuanlan.zhihu.com/p/538610416
Multi-Task Dense Retrieval via Model Uncertainty Fusion for Open-Domain Question Answering

使用CLIP Latent的分层文本条件的图像生成
https://zhuanlan.zhihu.com/p/538403355
Hierarchical Text-Conditional Image Generation with CLIP Latents

RNG-KBQA: 用于知识库问答的生成增强型迭代排名
https://zhuanlan.zhihu.com/p/535420059
RNG-KBQA: Generation Augmented Iterative Ranking for Knowledge Base Question Answering

ViLT：没有卷积或区域监督的视觉和语言transformer
https://zhuanlan.zhihu.com/p/537416032
ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision

图像-文本匹配的相似性推理和过滤
https://zhuanlan.zhihu.com/p/537392389
Similarity Reasoning and Filtration for Image-Text Matching

关于图像-文本检索的可复现性问题
https://zhuanlan.zhihu.com/p/535969785
Where Does the Performance Improvement Come From? - A Reproducibility Concern about Image-Text Retrieval

多模态检索的跨语言跨模态预训练
https://zhuanlan.zhihu.com/p/535949214
Cross-lingual Cross-modal Pretraining for Multimodal Retrieval

检索和阅读：关于开放域问答的综合调查
https://zhuanlan.zhihu.com/p/535315271
Retrieving and Reading ： A Comprehensive Survey on Open-domain Question Answering

扩散-LM改善可控文本的生成
https://zhuanlan.zhihu.com/p/532644454
Diffusion-LM Improves Controllable Text Generation

通过随机过程建立的语言模型
https://zhuanlan.zhihu.com/p/507834523
LANGUAGE MODELING VIA STOCHASTIC PROCESSES

UNITER: 通用图像-文本表示法学习
https://zhuanlan.zhihu.com/p/510622677
UNITER： UNiversal Image-TExt Representation Learning

重新审视无监督的关系抽取
https://zhuanlan.zhihu.com/p/527512757
Revisiting Unsupervised Relation Extraction

Hateful Memes 多模态数据集
https://zhuanlan.zhihu.com/p/509654285
The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes

OCR-VQA: 通过阅读图像中的文字进行可视化答题
https://zhuanlan.zhihu.com/p/506453197
OCR-VQA： Visual Question Answering by Reading Text in Images

PaLM：大模型的规模探索
https://zhuanlan.zhihu.com/p/503968575
PaLM： Scaling Language Modeling with Pathways

PICARD: 文本到SQL的自回归解码语言模型
https://zhuanlan.zhihu.com/p/504133233
PICARD:Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models

Survey：复杂知识库问答
https://zhuanlan.zhihu.com/p/503965660
A Survey on Complex Knowledge Base Question Answering： Methods, Challenges and Solutions

Tweets的多模态实体链接
https://zhuanlan.zhihu.com/p/502269546
Multimodal Entity Linking for Tweets

ELQ: 高效的一次性端到端实体链接的问题
https://zhuanlan.zhihu.com/p/497708749
Efficient One-Pass End-to-End Entity Linking for Questions

可扩展的zero-shot实体链接与密集实体检索
https://zhuanlan.zhihu.com/p/495291925
Scalable Zero-shot Entity Linking with Dense Entity Retrieval

实体链接技术和解决方案
https://zhuanlan.zhihu.com/p/495278919
Entity Linking Meets Deep Learning： Techniques and Solutions

多模态实体链接: 一个新的数据集和一个基线
https://zhuanlan.zhihu.com/p/494415154
Multimodal Entity Linking：A New Dataset and A Baseline

Zeroshot多模态命名实体歧义的社交媒体帖子
https://zhuanlan.zhihu.com/p/494350411
Zeroshot Multimodal Named Entity Disambiguation for Noisy Social Media Posts

Survey:多模态知识图谱的构建和应用
https://zhuanlan.zhihu.com/p/491610188
Multi-Modal Knowledge Graph Construction and Application： A Survey

多模态知识图谱完成
https://zhuanlan.zhihu.com/p/490909554
Embedding Multimodal Relational Data for Knowledge Base Completion

从Tweets中建立一个多模态实体链接数据集
https://zhuanlan.zhihu.com/p/490888754
Building a Multimodal Entity Linking Dataset From Tweets

ConVSE：视觉-语义嵌入的对比性学习
https://zhuanlan.zhihu.com/p/490315692
Contrastive Learning of Visual-Semantic Embeddings

用于图像-文本匹配的视觉语义推理
https://zhuanlan.zhihu.com/p/490302244
Visual Semantic Reasoning for Image-Text Matching

CLIP:从自然语言监督中学习可迁移的视觉模型
https://zhuanlan.zhihu.com/p/478889210
Learning Transferable Visual Models From Natural Language Supervision

VisualSem: 用于视觉和语言的高质量知识图谱
https://zhuanlan.zhihu.com/p/478679587
VisualSem： A High-quality Knowledge Graph for Vision & Language

MET:用多模态知识库进行多模态实体标注
https://zhuanlan.zhihu.com/p/478627581
Multimodal Entity Tagging with Multimodal Knowledge Base

M4C多模态transformer对TextVQA进行迭代式答案预测
https://zhuanlan.zhihu.com/p/477062474
Iterative Answer Prediction with Pointer-Augmented Multimodal Transformers for TextVQA

TPLinker: 通过token对链接的单阶段联合提取实体和关系
https://zhuanlan.zhihu.com/p/471975897
TPLinker：Single-stage Joint Extraction of Entities and Relations Through Token Pair Linking

UNIRE：实体关系抽取的统一标签空间
https://zhuanlan.zhihu.com/p/454398188
UNIRE： A Unified Label Space for Entity Relation Extraction

SCIERC实体识别和关系抽取的英文数据集
https://zhuanlan.zhihu.com/p/462638191
Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Construction

SHAP解释树模型
https://zhuanlan.zhihu.com/p/459470781
A Unified Approach to Interpreting Model Predictions

HySPA：用于可扩展的文本到图提取的混合跨度生成
https://zhuanlan.zhihu.com/p/454339907
HySPA： Hybrid Span Generation for Scalable Text-to-Graph Extraction

对树模型进行局部解释
https://zhuanlan.zhihu.com/p/458958125
From local explanations to global understanding with explainable AI for trees

XGBoost 一个可扩展的tree boosting系统
https://zhuanlan.zhihu.com/p/459470547
XGBoost： A Scalable Tree Boosting System

用图卷积网络进行联合信息提取
https://zhuanlan.zhihu.com/p/454304430
Cross-Task Instance Representation Interactions and Label Dependencies for Joint Information Extraction with Graph Convolutional Networks

用全局特征进行信息提取的联合神经模型
https://zhuanlan.zhihu.com/p/454108143
A Joint Neural Model for Information Extraction with Global Features

利用打包浮动标记进行实体和关系抽取
https://zhuanlan.zhihu.com/p/454398356
Pack Together： Entity and Relation Extraction with Levitated Marker

社媒的热点主题预测
https://zhuanlan.zhihu.com/p/453417760
Hot topic prediction considering influence and expertise in social media

社交媒体大数据分析
https://zhuanlan.zhihu.com/p/453838204
Big Social Media Data Analytics

PRCA和IGA联合建模分析顾客满意度
https://zhuanlan.zhihu.com/p/449441801
Integrating methods for the prioritization of innovations and improvements in services

基于实体相对位置表示的多头选择的联合实体和关系抽取
https://zhuanlan.zhihu.com/p/448112834
Entity Relative Position Representation based Multi-head Selection for Joint Entity and Relation Extraction

Kano加诺模型与数据挖掘的整合来预测客户满意度
https://zhuanlan.zhihu.com/p/448486378
Concept Paper Kano Model Integration with Data Mining to Predict Customer Satisfaction

PRCA惩罚-奖励-对比分析：在顾客满意度研究中的应用
https://zhuanlan.zhihu.com/p/449242532
Penalty–Reward-Contrast Analysis： a review of its application in customer satisfaction research

极端多标签文本分类的快速多分辨率transformer微调技术
https://zhuanlan.zhihu.com/p/445661903
Fast Multi-Resolution Transformer Fine-tuning for Extreme Multi-label Text Classification

多任务学习加强多标签文本分类
https://zhuanlan.zhihu.com/p/445661700
Enhancing Label Correlation Feedback in Multi-Label Text Classification via Multi-Task Learning

基于TFIDF和GloVe的多标签文本分类
https://zhuanlan.zhihu.com/p/445587449
Deep Learning Based Multi-Label Text Classification of UNGA Resolutions

动态语义表示和深度神经网络结合的多标签文本分类方法
https://zhuanlan.zhihu.com/p/445517482
A multi-label text classification method via dynamic semantic representation model and deep neural network

表-序列编码器联合提取实体和关系
https://zhuanlan.zhihu.com/p/440722315
Two are Better than One：Joint Entity and Relation Extraction with Table-Sequence Encoders

用上下文跨度表示的实体、关系和事件提取
https://zhuanlan.zhihu.com/p/443573825
Entity, Relation, and Event Extraction with Contextualized Span Representations

多头选择框架的BERT联合实体关系抽取
https://zhuanlan.zhihu.com/p/443577609
BERT-Based Multi-Head Selection for Joint Entity-Relation Extraction

实体和关系抽取的简单方法
https://zhuanlan.zhihu.com/p/440704543
A Frustratingly Easy Approach for Entity and Relation Extraction

BenchIE：基于事实而非token的开放式信息提取评估
https://zhuanlan.zhihu.com/p/438437407
BenchIE： Open Information Extraction Evaluation Based on Facts, Not Tokens

OpenIE6：用于开放信息提取的迭代网格标签和协调分析
https://zhuanlan.zhihu.com/p/438007291
OpenIE6: Iterative Grid Labeling and Coordination Analysis for Open Information Extraction

医学放射科报告生成与知识图谱
https://zhuanlan.zhihu.com/p/436319124
When Radiology Report Generation Meets Knowledge Graph

BUTD：自下而上和自上而下的注意力多模态模型
https://zhuanlan.zhihu.com/p/435174845
Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering

UniT：多模态多任务模型
https://zhuanlan.zhihu.com/p/434243735
UniT: Multimodal Multitask Learning with a Unified Transformer

ACT: 自适应聚类transformer端到端目标检测
https://zhuanlan.zhihu.com/p/435175009
End-to-End Object Detection with Adaptive Clustering Transformer

VisualBert: 经过预训练的多模态模型
https://zhuanlan.zhihu.com/p/434272329
VISUALBERT: A SIMPLE AND PERFORMANT BASELINE FOR VISION AND LANGUAGE

双线性注意力网络（多模态）
https://zhuanlan.zhihu.com/p/432970660
Bilinear Attention Networks

OpenUE：从文本中提取通用信息的开放工具箱
https://zhuanlan.zhihu.com/p/431805279
OpenUE: An Open Toolkit of Universal Extraction from Text

多模态分类的跨模态检索增强功能
https://zhuanlan.zhihu.com/p/432389016
Cross-Modal Retrieval Augmentation for Multi-Modal Classification

艺术品类图像的双流多模态模型情感分析
https://zhuanlan.zhihu.com/p/432373663
Understanding of Emotion Perception from Art

基于远端监督的开放领域数据的命名实体识别
https://zhuanlan.zhihu.com/p/428925959
Named Entity Recognition for Open Domain Data Based on Distant Supervision

MISA: 多模态情感分析的模态不变和模态特定表示
https://zhuanlan.zhihu.com/p/430407430
MISA: Modality-Invariant and -Specific Representations for Multimodal Sentiment Analysis

利用网络资源发现新的实体
https://zhuanlan.zhihu.com/p/428861588
Emerging Entity Discovery Using Web Sources

知识图谱完成方法的重新评估
https://zhuanlan.zhihu.com/p/428088532
A Re-evaluation of Knowledge Graph Completion Methods

Info-HCVAE问答对生成
https://zhuanlan.zhihu.com/p/421265798
Generating Diverse and Consistent QA pairs from Contexts with Information-Maximizing Hierarchical Conditional VAEs

多任务学习框架下的观点三元组提取
https://zhuanlan.zhihu.com/p/426376153
A Multi-task Learning Framework for Opinion Triplet Extraction

ASAP: 中文评论数据集：aspect的情感分析
https://zhuanlan.zhihu.com/p/425981216
ASAP: A Chinese Review Dataset Towards Aspect Category Sentiment Analysis and Rating Prediction

DiaKG：用于构建医学知识图谱的糖尿病标注数据集
https://zhuanlan.zhihu.com/p/424733768
DiaKG: an Annotated Diabetes Dataset for Medical Knowledge Graph Construction

P-Tuning v2: 与微调性能相等的提示性优化
https://zhuanlan.zhihu.com/p/423902902
P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks

TechKG：一个大规模的中文技术导向的知识图谱
https://zhuanlan.zhihu.com/p/420557472
TechKG: A Large-Scale Chinese Technology-Oriented Knowledge Graph

知识图谱增强的aspect情感分析
https://zhuanlan.zhihu.com/p/414252384
Scalable End-to-End Training of Knowledge Graph-Enhanced Aspect Embedding for Aspect Level Sentiment Analysis

Pre-train, Prompt, and Predict: 自然语言处理中prompting方法总结
https://zhuanlan.zhihu.com/p/411341801
Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing

用常识性知识图谱进行zero-shot学习
https://zhuanlan.zhihu.com/p/410561852
Zero-Shot Learning with Common Sense Knowledge Graphs

用于自然语言推理的随机答案网络
https://zhuanlan.zhihu.com/p/409085184
Stochastic Answer Networks for Natural Language Inference

用于自然语言理解的微软多任务深度神经网络工具包
https://zhuanlan.zhihu.com/p/408851910
The Microsoft Toolkit of Multi-Task Deep Neural Networks for Natural Language Understanding

大型神经语言模型的对抗性训练
https://zhuanlan.zhihu.com/p/408582923
Adversarial Training for Large Neural Language Models

行为克隆强化学习玩CS反恐精英
https://zhuanlan.zhihu.com/p/403123868
Counter-Strike Deathmatch with Large-Scale Behavioural Cloning

HyperTools: 可视化和操作高维度据的Python工具箱
https://zhuanlan.zhihu.com/p/407691325
HyperTools: A Python toolbox for visualizing and manipulating high-dimensional data

显存优化
https://zhuanlan.zhihu.com/p/407429742
Training Deep Nets with Sublinear Memory Cost

自然语言查询问答模型
https://zhuanlan.zhihu.com/p/406453009
Database Reasoning Over Text

Transformer Survey
https://zhuanlan.zhihu.com/p/405623198
A Survey of Transformers

X-modaler: 用于跨模态分析的多功能和高性能的代码库
https://zhuanlan.zhihu.com/p/402620759
X-modaler: A Versatile and High-performance Codebase for Cross-modal Analytics

PonderNet：适应性模型，提高模型计算效率
https://zhuanlan.zhihu.com/p/401874414
PonderNet: Learning to Ponder

模式引导下的多领域对话数据集
https://zhuanlan.zhihu.com/p/401779764
Towards Scalable Multi-Domain Conversational Agents: The Schema-Guided Dialogue Dataset

新冠COVID-19文献知识图谱构建
https://zhuanlan.zhihu.com/p/400944819
COVID-19 Literature Knowledge Graph Construction and Drug Repurposing Report Generation

AutoKnow: 上千种产品的自动知识收集
https://zhuanlan.zhihu.com/p/399419662
AutoKnow: Self-Driving Knowledge Collection for Products of Thousands of Types

用于联合意向分类和槽位填充的BERT
https://zhuanlan.zhihu.com/p/399103189
BERT for Joint Intent Classification and Slot Filling

数据标注的质量控制案例:TDT语料
https://zhuanlan.zhihu.com/p/398515851
Quality Control in Large Annotation Projects Involving Multiple Judges: The Case of the TDT Corpora

DeiT: 蒸馏的图像transformer模型
https://zhuanlan.zhihu.com/p/394627382
Training data-efficient image transformers & distillation through attention

ViT: transformer用于图像识别
https://zhuanlan.zhihu.com/p/394288661
AN IMAGE IS WORTH 16X16 WORDS: TRANSFORMERS FOR IMAGE RECOGNITION AT SCALE

YOLOX
https://zhuanlan.zhihu.com/p/393955749
YOLOX: Exceeding YOLO Series in 2021

完善知识图谱总结
https://zhuanlan.zhihu.com/p/393885109
Knowledge Graph Refinement: A Survey of Approaches and Evaluation Methods

AdaShare: 高效的深度多任务学习
https://zhuanlan.zhihu.com/p/393243186
AdaShare: Learning What To Share For Efficient Deep Multi-Task Learning

知识图谱总结： 表示、获取和应用
https://zhuanlan.zhihu.com/p/392429070
A Survey on Knowledge Graphs: Representation, Acquisition and Applications

REPAINT：深度强化学习中的知识迁移
https://zhuanlan.zhihu.com/p/391824772
REPAINT: Knowledge Transfer in Deep Reinforcement Learning

重新审视Rainbow
https://zhuanlan.zhihu.com/p/391126427
Revisiting Rainbow: Promoting more Insightful and Inclusive Deep Reinforcement Learning Research

遗传进化强化学习算法
https://zhuanlan.zhihu.com/p/389895408
EVOLVING REINFORCEMENT LEARNING ALGORITHMS

强化学习环境：Google足球游戏
https://zhuanlan.zhihu.com/p/389567798
Google Research Football: A Novel Reinforcement Learning Environment

无监督文本摘要
https://zhuanlan.zhihu.com/p/388911963
Learning to Encode Text as Human-Readable Summaries using Generative Adversarial Networks

ERNIE 3.0: 用于语言理解和生成的大规模知识强化预训练
https://zhuanlan.zhihu.com/p/388172601
ERNIE 3.0: LARGE-SCALE KNOWLEDGE ENHANCED PRE-TRAINING FOR LANGUAGE UNDERSTANDING AND GENERATION

COMET：自动构建知识图谱的常识transformer
https://zhuanlan.zhihu.com/p/388106049
COMET : Commonsense Transformers for Automatic Knowledge Graph Construction

D2S: 通过基于查询的文本总结进行文档到幻灯片的生成
https://zhuanlan.zhihu.com/p/387544973
D2S: Document-to-Slide Generation Via Query-Based Text Summarization

斗地主强化学习
https://zhuanlan.zhihu.com/p/385496621
DouZero: Mastering DouDizhu with Self-Play Deep Reinforcement Learning

通过知识蒸馏改进多任务深度神经网络以促进自然语言理解
https://zhuanlan.zhihu.com/p/384120253
Improving Multi-Task Deep Neural Networks via Knowledge Distillation for Natural Language Understanding

使用深度强化学习玩MOBA游戏
https://zhuanlan.zhihu.com/p/378789632
Towards Playing Full MOBA Games with Deep Reinforcement Learning

MOBA游戏的复杂控制与深度强化学习
https://zhuanlan.zhihu.com/p/379091485
Mastering Complex Control in MOBA Games with Deep Reinforcement Learning

用强化学习玩英雄联盟
https://zhuanlan.zhihu.com/p/363495437
Deep Learning Bot for League of Legends

ATARI游戏的Model based的强化学习
https://zhuanlan.zhihu.com/p/363279136
MODEL BASED REINFORCEMENT LEARNING FOR ATARI

关于视频游戏的深度强化学习算法
https://zhuanlan.zhihu.com/p/363115461
A Survey of Deep Reinforcement Learning in Video Games

用于自然语言理解的多任务深度神经网络
https://zhuanlan.zhihu.com/p/383137481
Multi-Task Deep Neural Networks for Natural Language Understanding

模仿学习: 自动排序的演示学习
https://zhuanlan.zhihu.com/p/382272429
Better-than-Demonstrator Imitation Learning via Automatically-Ranked Demonstrations

预训练编码器文本摘要
https://zhuanlan.zhihu.com/p/381490918
Text Summarization with Pretrained Encoders

Survey: 多任务学习
https://zhuanlan.zhihu.com/p/381229374
A Survey on Multi-Task Learning

DDPG论文: 深强化学习连续控制
https://zhuanlan.zhihu.com/p/371451813
CONTINUOUS CONTROL WITH DEEP REINFORCEMENT LEARNING

使用图像做无地图导航的强化学习
https://zhuanlan.zhihu.com/p/379270657
Using RGB Image as Visual Input for Mapless Robot Navigation

Pettingzoo：类似gym的多Agent强化学习的环境
https://zhuanlan.zhihu.com/p/375049925
PettingZoo: Gym for Multi-Agent Reinforcement Learning

RIIT: 最新的多Agent合作控制强化学习算法
https://zhuanlan.zhihu.com/p/368284926
RIIT: Rethinking the Importance of Implementation Tricks in Multi-Agent Reinforcement Learning

复现深度强化学习算法效果的因素
https://zhuanlan.zhihu.com/p/377369590
Deep Reinforcement Learning that Matters

PPO算法
https://zhuanlan.zhihu.com/p/376978985
Proximal Policy Optimization Algorithms

交互式的强化学习
https://zhuanlan.zhihu.com/p/379871647
REINFORCEMENT LEARNING WITH HUMAN ADVICE: A SURVEY

半监督机器翻译的简单基准
https://zhuanlan.zhihu.com/p/378838885
A Simple Baseline to Semi-Supervised Domain Adaptation for Machine Translation

Survey: 深度神经网络翻译
https://zhuanlan.zhihu.com/p/378524968
A Survey of Deep Learning Techniques for Neural Machine Translation

MMBT: 用于图像和文本分类的有监督多模态双向Transformer
https://zhuanlan.zhihu.com/p/373581881
Supervised Multimodal Bitransformers for Classifying Images and Text

XLM-R: 大规模无监督跨语言表示模型
https://zhuanlan.zhihu.com/p/372978148
Unsupervised Cross-lingual Representation Learning at Scale

跨语言的语言模型预训练
https://zhuanlan.zhihu.com/p/372001934
Cross-lingual Language Model Pretraining

神经机器翻译的无监督领域适应
https://zhuanlan.zhihu.com/p/371626610
Unsupervised Domain Adaptation for Neural Machine Translation with Domain-Aware Feature Embeddings

域适应翻译中的单词表适应方法
https://zhuanlan.zhihu.com/p/371392857
Vocabulary Adaptation for Domain Adaptation in Neural Machine Translation

使用术语限制的神经网络翻译
https://zhuanlan.zhihu.com/p/370661928
Training Neural Machine Translation To Apply Terminology Constraints

Survey: 机器翻译的领域适应和多领域适应
https://zhuanlan.zhihu.com/p/370390321
Domain Adaptation and Multi-Domain Adaptation for Neural Machine Translation: A Survey

M2M-100: 多语言翻译模型
https://zhuanlan.zhihu.com/p/368226087
Beyond English-Centric Multilingual Machine Translation

FAIRSEQ 语音到文本模型
https://zhuanlan.zhihu.com/p/361585021
FAIRSEQ S2T: Fast Speech-to-Text Modeling with FAIRSEQ

自然语言中的强化学习
https://zhuanlan.zhihu.com/p/364138298
A Survey of Reinforcement Learning Informed by Natural Language

MAAC注意力的演员评论家: Multi-Agent强化学习
https://zhuanlan.zhihu.com/p/366413456
Actor-Attention-Critic for Multi-Agent Reinforcement Learning

mBART：多语言翻译预训练模型
https://zhuanlan.zhihu.com/p/366525006
Multilingual Denoising Pre-training for Neural Machine Translation

MobileBERT:用于资源限制设备的紧凑型任务型BERT
https://zhuanlan.zhihu.com/p/365329984
MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices

更快的深度自适应transformers
https://zhuanlan.zhihu.com/p/364807276
Faster Depth-Adaptive Transformers

利用远端监督的强化学习关系抽取
https://zhuanlan.zhihu.com/p/364444877
Large Scaled Relation Extraction with Reinforcement Learning

GLRE模型文档级关系抽取
https://zhuanlan.zhihu.com/p/360980109
Global-to-Local Neural Networks for Document-Level Relation Extraction

关系抽取的注意力引导图卷积网络
https://zhuanlan.zhihu.com/p/357518473
Attention Guided Graph Convolutional Networks for Relation Extraction

关系抽取Review（附上中文关系抽取的数据及代码)
https://zhuanlan.zhihu.com/p/356551233
More Data, More Relations, More Context and More Openness: A Review and Outlook for Relation Extraction

DocRED: 大型文档级关系抽取数据集
https://zhuanlan.zhihu.com/p/356077381
DocRED: A Large-Scale Document-Level Relation Extraction Dataset

文档级关系抽取：图增强双重注意力网络
https://zhuanlan.zhihu.com/p/355473773
Graph Enhanced Dual Attention Network for Document-Level Relation Extraction

SENTIX:跨领域情感分析预训练模型
https://zhuanlan.zhihu.com/p/350924103
SENTIX: A Sentiment-Aware Pre-Trained Model for Cross-Domain Sentiment Analysis

DEBERTA：解耦注意力的解码增强型BERT
https://zhuanlan.zhihu.com/p/348704980
DEBERTA: DECODING-ENHANCED BERT WITH DISENTANGLED ATTENTION

SentiBERT： 基于可迁移的transformer的组合的情感语义预训练模型
https://zhuanlan.zhihu.com/p/347854488
SentiBERT: A Transferable Transformer-Based Architecture for Compositional Sentiment Semantics

SentiLARE：带有语言知识的情感感知预训练模型
https://zhuanlan.zhihu.com/p/346202158
SentiLARE: Sentiment-Aware Language Representation Learning with Linguistic Knowledge

SpanBERT：通过表示和预测跨度来改善预训练的模型(2020年1月修订)
https://zhuanlan.zhihu.com/p/345401994
SpanBERT: Improving Pre-training by Representing and Predicting Spans

抱抱脸🤗Transformers论文(2020年)
https://zhuanlan.zhihu.com/p/344553832
Transformers: State-of-the-Art Natural Language Processing

SentencePiece:子词tokenizer和detokenizer(2019年12月更新)
https://zhuanlan.zhihu.com/p/343634730
SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing

ALBERT: 精简的BERT自监督的语言表示模型(2020年2月更新)
https://zhuanlan.zhihu.com/p/343426088
ALBERT: A LITE BERT FOR SELF-SUPERVISED LEARNING OF LANGUAGE REPRESENTATIONS

上下文和实体名称哪个对关系抽取更重要(2020年12月论文)
https://zhuanlan.zhihu.com/p/342360873
Learning from Context or Names? An Empirical Study on Neural Relation Extraction

Google Big Bird：长序列的transformers(2020年论文)
https://zhuanlan.zhihu.com/p/342005602
Big Bird: Transformers for Longer Sequences

mT5: 多国语言版T5(中文T5)(2020年10月论文)
https://zhuanlan.zhihu.com/p/340288423
mT5: A massively multilingual pre-trained text-to-text transformer

Google T5: 统一文本到文本迁移学习研究 (2020年7月论文)-Part3
https://zhuanlan.zhihu.com/p/339502041
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer

CharBERT：字符感知的预训练语言模型(2020年11月论文)
https://zhuanlan.zhihu.com/p/337587788
CharBERT: Character-aware Pre-trained Language Model

AdaBERT：可导神经结构搜索的任务自适应BERT压缩(2020年1月论文)
https://zhuanlan.zhihu.com/p/337305614
AdaBERT: Task-Adaptive BERT Compression with Differentiable Neural Architecture Search

EasyTransfer–阿里NLP深度迁移学习平台(2020年11月论文)
https://zhuanlan.zhihu.com/p/336730123
EasyTransfer – A Simple and Scalable Deep Transfer Learning Platform for NLP Applications

ConvBERT: 基于跨度的动态卷积BERT(2020年11月论文)
https://zhuanlan.zhihu.com/p/336409975
ConvBERT: Improving BERT with Span-based Dynamic Convolution

MacBERT: 中文自然语言预训练模型(2020年11月论文)
https://zhuanlan.zhihu.com/p/333202482
Revisiting Pre-trained Models for Chinese Natural Language Processing

FLAT: 使用Flat-Lattice Transformer结构中文NER(2020年5月论文)
https://zhuanlan.zhihu.com/p/326135985
FLAT: Chinese NER Using Flat-Lattice Transformer

ELECTRA: 区别于BERT，使用判别器构建预训练模型（2020年3月论文)
https://zhuanlan.zhihu.com/p/323931207
ELECTRA: PRE-TRAINING TEXT ENCODERS AS DISCRIMINATORS RATHER THAN GENERATORS

无监督的深度嵌入式聚类(2016年论文)
https://zhuanlan.zhihu.com/p/313662693
Unsupervised Deep Embedding for Clustering Analysis

BOND：半监督的BERT开放域命名实体识别(2020年6月论文)
https://zhuanlan.zhihu.com/p/307454757
BOND: BERT-Assisted Open-Domain Named Entity Recognition with Distant Supervision

使用半监督和监督学习检测虚假的在线评论(2019年2月)
https://zhuanlan.zhihu.com/p/301268523
Detection of fake online reviews using semi-supervised and supervised learning

垃圾观点检测：使用基于多次迭代的图模型(2020年论文)
https://zhuanlan.zhihu.com/p/300841251
Opinion spam detection: Using multi-iterative graph-based model

基于预训练和序列迁移的语法纠错系统(2019年7月)
https://zhuanlan.zhihu.com/p/288219713
A Neural Grammatical Error Correction System Built On Better Pre-training and Sequential Transfer Learning

序列到序列的中文语法纠错(2018年)
https://zhuanlan.zhihu.com/p/285211193
A Sequence to Sequence Learning for Chinese Grammatical Error Correction

BERT-of-Theseus: 通过逐步更换模块压缩BERT模型(2020年10月)
https://zhuanlan.zhihu.com/p/283118184
BERT-of-Theseus: Compressing BERT by Progressive Module Replacing

TextBrewer：用于自然语言处理的开源知识蒸馏工具包(2020.04)
https://zhuanlan.zhihu.com/p/275722016
TextBrewer: An Open-Source Knowledge Distillation Toolkit for Natural Language Processing

TinyBERT(华为)：自然语言理解之BERT蒸馏(2020.10)
https://zhuanlan.zhihu.com/p/273467698
TinyBERT: Distilling BERT for Natural Language Understanding

BERT-PKD: BERT模型压缩之耐心知识蒸馏（2019.08）
https://zhuanlan.zhihu.com/p/274329168
Patient Knowledge Distillation for BERT Model Compression

distilBert: Bert 蒸馏到简单的BiLSTM(2019.03)
https://zhuanlan.zhihu.com/p/273543240
Distilling Task-Specific Knowledge from BERT into Simple Neural Networks

可转移域的端到端选择性对抗学习的Aspect-based情感分析(2019年10)
https://zhuanlan.zhihu.com/p/268320982
Transferable End-to-End Aspect-based Sentiment Analysis with Selective Adversarial Learning

联合Aspect-Sentiment主题嵌入的弱监督的情感分析(2020年10)
https://zhuanlan.zhihu.com/p/267744626
Weakly-Supervised Aspect-Based Sentiment Analysis via Joint Aspect-Sentiment Topic Embedding

观点目标提取和情感预测的统一模型(2019年)
https://zhuanlan.zhihu.com/p/268580604
A Unified Model for Opinion Target Extraction and Target Sentiment Prediction

用于目标情感分类的注意力编码器网络(2019年04）
https://zhuanlan.zhihu.com/p/270374318
Attentional Encoder Network for Targeted Sentiment Classification

评测BERT类细粒度情感分类的语言表示模型(2020.05)
https://zhuanlan.zhihu.com/p/268012476
Language Representation Models for Fine-Grained Sentiment Classification

利用BERT进行端到端aspect-based的情感分析(2019年10）
https://zhuanlan.zhihu.com/p/268801608
Exploiting BERT for End-to-End Aspect-based Sentiment Analysis

Aspect-level基于注意力的LSTM 情感分类(2016年)
https://zhuanlan.zhihu.com/p/267254311
Attention-based LSTM for Aspect-level Sentiment Classification

BERT-EMD：多层对多层映射的BERT蒸馏（2020年10月）
https://zhuanlan.zhihu.com/p/266602585
BERT-EMD: Many-to-Many Layer Mapping for BERT Compression with Earth Mover’s Distance

媲美GPT-3的变种PET模型(2020年9月论文)
https://zhuanlan.zhihu.com/p/265646470
It’s Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners

PRADO: 移动设备上的投影注意力文本分类网络(2019年论文)
https://zhuanlan.zhihu.com/p/265042724
PRADO: Projection Attention Networks for Document Classification On-Device

图神经网络的图归一化(2020.09论文)
https://zhuanlan.zhihu.com/p/260811611
Learning Graph Normalization for Graph Neural Network

基于无标签数据Copy-Augmented预训练结构改善语法纠错(2019.06)
https://zhuanlan.zhihu.com/p/258091623
Improving Grammatical Error Correction via Pre-Training a Copy-Augmented Architecture with Unlabeled Data

多层卷积编解码器神经网络语法纠错(2018.01论文)
https://zhuanlan.zhihu.com/p/248000441
A Multilayer Convolutional Encoder-Decoder Neural Network for Grammatical Error Correction

近域微调的图片表格检测模型
https://zhuanlan.zhihu.com/p/248393029
The Benefits of Close-Domain Fine-Tuning for Table Detection in Document Images

IBM基于图片格式的表格识别(2020年03月论文)
https://zhuanlan.zhihu.com/p/245032050
Image-based table recognition: data, model, and evaluation

FASPell基于DAE解码器的Spell Checker（2019.09）
https://zhuanlan.zhihu.com/p/231626818
FASPell: A Fast, Adaptable, Simple, Powerful Chinese Spell Checker Based On DAE-Decoder Paradigm

个性化语法错误​​纠正（2020.06论文）
https://zhuanlan.zhihu.com/p/231190671
Personalizing Grammatical Error Correction: Adaptation to Proficiency Level and L1

Reformer: 搞笑（高效）的transformer结构(2020年2月Google)
https://zhuanlan.zhihu.com/p/208134502
REFORMER: THE EFFICIENT TRANSFORMER

2阶段中文纠错模型(2019论文)
https://zhuanlan.zhihu.com/p/199551915
A Two-Stage Model for Chinese Grammatical Error Correction

CLUENER2020 2020年汉语NER和Benchmark
https://zhuanlan.zhihu.com/p/197488236
CLUENER2020: FINE-GRAINED NAMED ENTITY RECOGNITION DATASET AND BENCHMARK FOR CHINESE

Google 最新 NLP语言模型可解释性可视化分析工具（2020-8月论文）
https://zhuanlan.zhihu.com/p/188617204
The Language Interpretability Tool: Extensible, Interactive Visualizations and Analysis for NLP Models

聊天机器人架构设计和开发2-架构和设计（论文By Jack Cahn )
https://zhuanlan.zhihu.com/p/181491658
CHATBOT: Architecture, Design, & Development By Jack Cahn

DocBert Bert用作文档分类(2019年8月论文)
https://zhuanlan.zhihu.com/p/180475198
DocBERT: BERT for Document Classification

TableBank：用于表检测和识别的基准数据集
https://zhuanlan.zhihu.com/p/170365926
TableBank: A Benchmark Dataset for Table Detection and Recognition

LayoutLM 微软预训练模型图片类文档分类和实体识别(2020年6月论文)
https://zhuanlan.zhihu.com/p/166128964
LayoutLM: Pre-training of Text and Layout for Document Image Understanding

UNILM 微软预训练的NLU和NLG结合模型(2019-10论文)
https://zhuanlan.zhihu.com/p/164736442
Unified Language Model Pre-training for Natural Language Understanding and Generation

DIET模型 rasa 聊天机器人核心模型论文（2020年5月论文)
https://zhuanlan.zhihu.com/p/162995854
Dual Intent and Entity Transformer

NLP之MixText 半监督文本分类(2020年4月论文解读)
https://zhuanlan.zhihu.com/p/156091468
MixText: Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text Classification

NLP 之数据增强算法(论文解读-2019年论文)
https://zhuanlan.zhihu.com/p/152633064
EDA: Easy Data Augmentation Techniques for Boosting Performance onText Classification Tasks

TFIDF+Wordembedding无监督多标签文本分类算法（论文解读)
https://zhuanlan.zhihu.com/p/152526817
Improving Recall and Precision in Unsupervised Multi-Label Document Classifification Tasks by Combining Word Embeddings with TF-IDF

评估对于少量样本使用Bert进行fine-tunning的优化方法（论文解读)
https://zhuanlan.zhihu.com/p/152523646
Revisiting Few-sample BERT Fine-tuning

SYNTHESIZER代替self-attention机制（Google论文解读)
https://zhuanlan.zhihu.com/p/152518921
Rethinking Self-Attention in Transformer Models