caiyinqiong

Welcome to my blog ~~

个人杂谈之SIGIR2020

(完成)SIGIR2020 (TKL)Local Self-Attention over Long Text for Efficient Document Retrieval 对Transformer-kernel的改进,主要是针对Transformer-kernel用于长文本时。之前的基于Transformer的模型用于长文本时通常的做法是直接截断...

个人杂谈之SIGIR2019

(完成)SIGIR2019(不包括openQA相关) (CQA中的重复问题检测)Adaptive Multi-Attention Network Incorporating Answer Information for Duplicate Question Detection 动机:paired answers can provide effecti...

个人杂谈之SIGIR2018

(完成)SIGIR2018(共录取86篇,录取率21%) (IR匹配)Modeling Diverse Relevance Patterns in Ad-hoc Retrieval(范意兴师兄) !!(排序)Learning a Deep Listwise Context Model for Ranking Refinement !!...

个人杂谈之ICLR2021

Universal Sentence Representations Learning with Conditional Masked Language Model A novel pre-training technique CMLM for unsupervised sentence representation learning on unlab...

个人杂谈之ICLR2019

(完成)ICLR2019(可能需要再滤一遍) (面向任务型对话系统)Global-to-local Memory Pointer Networks for Task-Oriented Dialogue 方法过于数学(对话系统的问题生成)Large-Scale Answerer in Questioner’s Mind for Visual Dialog Question Gene...

个人杂谈之ICLR2018

(完成)ICLR2018 (2.3%的口头展示,31.4%的poster接受,9%的workshop) !!!!(DIIN)Natural Language Inference over Interaction Space !!!!(IR,多任务学习)Multi-Task Learning for Document Ranking and Query ...

个人杂谈之EMNLP2020

(完成)EMNLP2020 AmbigQA: Answering Ambiguous Open-domain Questions 研究开放域问答中问题的歧义性,找到歧义性问题对应的多个答案,并针对每一个答案对问题进行重写以实现消歧。 提出了一个对应的任务和数据集,以及baseline。 【句子表示学习】An...

个人杂谈之ACL2020

(完成)ACL2020(571 篇长文和 208 篇短文) (效率)DeFormer: Decomposing Pre-trained Transformers for Faster Question Answering(解耦低层的BERT) (效率)DeeBERT: Dynamic Early Exiting for Accelerating BERT...

个人杂谈之ACL2019

(完成)ACL2019 RankQA: Neural Question Answering with Answer Re-Ranking Episodic Memory Reader: Learning What to Remember for Question Answering from Streaming Data Retrieve, Read, Rerank: Tow...

个人杂谈之ICLR2020

MRC Transformer-XH: Multi-hop question answering with eXtra Hop attention 内容简介:解决多跳推理的MRC问题。提出Transformer-XH结构,即每个passage在建模时不仅考虑自己的内部信息,还考虑了其他和它有连边的passage节点信息。各passage节点之间的连边...