All Projects → jingyihiter → myDuReader

jingyihiter / myDuReader

Licence: other
2018百度机器阅读理解竞赛

Programming Languages

python
139335 projects - #7 most used programming language
shell
77523 projects

Projects that are alternatives of or similar to myDuReader

KMRC-Papers
A list of recent papers regarding knowledge-based machine reading comprehension.
Stars: ✭ 40 (+33.33%)
Mutual labels:  machine-reading-comprehension
expmrc
ExpMRC: Explainability Evaluation for Machine Reading Comprehension
Stars: ✭ 58 (+93.33%)
Mutual labels:  machine-reading-comprehension
Bi Att Flow
Bi-directional Attention Flow (BiDAF) network is a multi-stage hierarchical process that represents context at different levels of granularity and uses a bi-directional attention flow mechanism to achieve a query-aware context representation without early summarization.
Stars: ✭ 1,472 (+4806.67%)
Mutual labels:  bidaf
question-answering
No description or website provided.
Stars: ✭ 32 (+6.67%)
Mutual labels:  bidaf
DuReader QANet BiDAF
Using QANet and BiDAF on DuReader datasets
Stars: ✭ 34 (+13.33%)
Mutual labels:  bidaf
Question-Answering-based-on-SQuAD
Question Answering System using BiDAF Model on SQuAD v2.0
Stars: ✭ 20 (-33.33%)
Mutual labels:  bidaf
Cluedatasetsearch
搜索所有中文NLP数据集,附常用英文NLP数据集
Stars: ✭ 2,112 (+6940%)
Mutual labels:  machine-reading-comprehension
LICS2021 MRC
2021语言与智能技术竞赛:机器阅读理解任务
Stars: ✭ 25 (-16.67%)
Mutual labels:  machine-reading-comprehension
MobileQA
离线端阅读理解应用 QA for mobile, Android & iPhone
Stars: ✭ 49 (+63.33%)
Mutual labels:  machine-reading-comprehension
machine-comprehension
MACHINE READING COMPREHENSION WITH SELF-MATCHING NETWORKS
Stars: ✭ 44 (+46.67%)
Mutual labels:  machine-reading-comprehension
unanswerable qa
The official implementation for ACL 2021 "Challenges in Information Seeking QA: Unanswerable Questions and Paragraph Retrieval".
Stars: ✭ 21 (-30%)
Mutual labels:  machine-reading-comprehension
strategy
Improving Machine Reading Comprehension with General Reading Strategies
Stars: ✭ 35 (+16.67%)
Mutual labels:  machine-reading-comprehension
verseagility
Ramp up your custom natural language processing (NLP) task, allowing you to bring your own data, use your preferred frameworks and bring models into production.
Stars: ✭ 23 (-23.33%)
Mutual labels:  machine-reading-comprehension
MSMARCO-MRC-Analysis
Analysis on the MS-MARCO leaderboard regarding the machine reading comprehension task.
Stars: ✭ 20 (-33.33%)
Mutual labels:  machine-reading-comprehension
KMRC-Research-Archive
🗂 Research about Knowledge-based Machine Reading Comprehension
Stars: ✭ 25 (-16.67%)
Mutual labels:  machine-reading-comprehension

2018百度机器阅读理解竞赛

竞赛简介

机器阅读理解(Machine Reading Comprehension)是指让机器阅读文本,然后回答和阅读内容相关的问题。阅读理解是自然语言处理和人工智能领域的重要前沿课题,对于提升机器智能水平、使机器具有持续知识获取能力具有重要价值,近年来受到学术界和工业界的广泛关注。

为了促进阅读理解技术发展,中国中文信息学会中国计算机学会百度公司联手举办“2018机器阅读理解技术竞赛”。竞赛将提供面向真实应用场景的大规模中文阅读理解数据集, 旨在为研究者提供学术交流平台,进一步提升阅读理解的研究水平, 推动语言理解和人工智能领域技术和应用的发展。竞赛将在第三届“语言与智能高峰论坛”举办技术交流和颁奖。诚邀学术界和工业界的研究者和开发者参加本次竞赛!

竞赛详情

1. 竞赛任务

对于给定问题_q_及其对应的文本形式的候选文档集合_D=d1, d2, ..., dn,要求参评阅读理解系统自动对问题及候选文档进行分析, 输出能够满足问题的文本答案a。目标是a能够正确、完整、简洁地回答问题_q

2. 数据简介

数据详情 本次竞赛数据集来自搜索引擎真实应用场景,其中的问题为百度搜索用户的真实问题,每个问题对应5个候选文档文本及人工整理的优质答案。

数据集共包含30万问题,包括27万的训练集,1万开发集和2万测试集。其中20万数据已在DuReader发布,包括18万训练集、1万的开发集和1万的测试集。这部分数据可自由下载(下载地址),供参赛者训练和测试使用。报名截止后,新增的10万数据集也将在数据下载区发布。

3. 评价方法

竞赛基于测试集的人工标注答案,采用ROUGH-LBLEU4作为评价指标,以ROUGH-L为主评价指标。针对是非及实体类型问题,对ROUGE-L和BLEU4评价指标进行了微调, 适当增加了正确识别是非答案类型及匹配实体的得分奖励, 一定程度上弥补传统ROUGE-L和BLEU4指标对是非和实体类型问题评价不敏感的问题。

*详细的评价指标说明参见数据集中包含的“评测细则“文档。

4. 基线系统

竞赛将提供两个开源的阅读理解基线系统,基线系统的实现及结果评价请参考:开源系统和数据集论文

主要模型

BiDAF+self attention

实验

最终排名

排行榜

最终榜单排名24位(╥╯^╰╥)

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].