SemEval-2021 Task 12: Learning with Disagreements

This repository contains the Post-Evaluation data for SemEval-2021 Task 12: Learning with Disagreement, a shared task on learning to classify with datasets containing disagreements. The aim of this shared task is to provide a unified testing framework for learning from disagreements using the best-known datasets containing information about disagreements for interpreting language and classifying images: 1. LabelMe-IC: Image Classification using a subset of LabelMe images (Russell et al., 2008), is a widely used, community-created image classification dataset where images are assigned to one of 8 categories: highway, inside city, tall building, street, forest, coast, mountain, open country. Rodrigues and Pereira (2017) collected crowd labels for these images using Amazon Mechanical Turk (AMT). 2. CIFAR10-IC: Image Classification using a subset of CIFAR-10 dataset, https://www.cs.toronto.edu/~kriz/cifar.html. The entire dataset consists of colour images in 10 categories (airplane, automobile, bird, cat, deer, dog, frog, horse, ship, and truck). Crowdsourced labels for this dataset were collected by Peterson et al (2019). 3. PDIS: Information Status Classification using Phrase Detectives Information. Information Status Classification (IS) in Phrase Detectives (Poesio et al., 2019) dataset involves identifying the information status of a noun phrase: whether that noun phrase refers to new information or to old information. 4. Gimpel-POS: Part-of-Speech tagging using the Gimpel dataset (Gimpel et al., 2011) for Twitter posts. Plank et al.(2014b) mapped the Gimpel tags to the universal tag set (Petrov et al., 2011), using these tags as gold, and collected crowdsourced labels. 5. Humour: ranking one-line texts using pairwise funniness judgements (Simpson et al., 2019). Crowdworkers have annotated pairs of puns to indicate which is funniest. A gold standard ranking was produced using a large number of redundant annotations. The goal is to infer the gold standard ranking from a reduced number of crowdsourced judgements. <br> The files contained in this data collection are as follows:<br> starting_kit.zip - Base models used provided for the shared task. <br> practice_phase_data.zip - The training and development data used during the Practice Phase of the competition. <br> test_phase_data.zip - The test data, used during the Evaluation Phase of the competition Details of format of each dataset for each task can be found on Codalab.

本仓库收录SemEval-2021任务12「带分歧学习（Learning with Disagreement）」的赛后评估数据，该共享任务旨在针对包含标注分歧的数据集构建分类学习模型。本共享任务的核心目标是为从标注分歧数据中学习提供统一的测试框架，所用数据集为当前涵盖语言理解与图像分类两类任务的主流带标注分歧信息的权威数据集，具体包括以下5项：
1. LabelMe-IC：基于LabelMe图像子集构建的图像分类数据集（Russell等，2008），是广泛使用的社区共创图像分类数据集，每张图像被划分为8个类别之一：高速公路、城市内部、高层建筑、街道、森林、海岸、山地、开阔乡村。Rodrigues与Pereira（2017）通过亚马逊机械Turk（Amazon Mechanical Turk, AMT）为该数据集收集了众包标注。
2. CIFAR10-IC：基于CIFAR-10数据集子集构建的图像分类数据集，数据集官方网址为https://www.cs.toronto.edu/~kriz/cifar.html。完整数据集包含10个类别的彩色图像：飞机、汽车、鸟类、猫、鹿、狗、青蛙、马、船舶与卡车。Peterson等（2019）为该数据集收集了众包标注。
3. PDIS：基于《短语侦探》（Phrase Detectives）数据集的信息状态分类任务。《短语侦探》（Poesio等，2019）中的信息状态分类（Information Status Classification, IS）任务需要识别名词短语的信息状态，即该名词短语指代的是新信息还是已有信息。
4. Gimpel-POS：针对Twitter（推特）文本的词性标注任务，所用数据集为Gimpel数据集（Gimpel等，2011）。Plank等（2014b）将Gimpel标注集映射至通用标注集（universal tag set, Petrov等，2011），并以该通用标注集作为金标准，同时收集了众包标注。
5. Humour：基于成对趣味性评判的单行文本排序任务（Simpson等，2019）。众包工作者对双关语对进行标注，以指明哪一个更有趣。研究人员通过大量冗余标注构建了金标准排序结果，任务目标是从少量众包评判结果中推断出该金标准排序。

本数据集集合包含的文件如下：
- starting_kit.zip：为本次共享任务提供的基础模型工具包
- practice_phase_data.zip：大赛练习阶段所用的训练与开发集数据
- test_phase_data.zip：大赛评估阶段所用的测试集数据
各任务数据集的具体格式细节可在Codalab平台查询。