Repository for the Bias Benchmark for QA dataset.

Related tags

Deep LearningBBQ
Overview

BBQ

Repository for the Bias Benchmark for QA dataset.

Authors: Alicia Parrish, Angelica Chen, Nikita Nangia, Vishakh Padmakumar, Jason Phang, Jana Thompson, Phu Mon Htut, and Samuel R. Bowman.

About BBQ

It is well documented that NLP models learnsocial biases present in the world, but littlework has been done to show how these biasesmanifest in actual model outputs for appliedtasks like question answering (QA). We introduce the Bias Benchmark for QA (BBQ), adataset consisting of question-sets constructedby the authors that highlightattestedsocialbiases against people belonging to protectedclasses along nine different social dimensionsrelevant for U.S. English-speaking contexts.Our task evaluates model responses at two distinct levels: (i) given an under-informative context, test how strongly model answers reflectsocial biases, and (ii) given an adequately informative context, test whether the model’s biases still override a correct answer choice. Wefind that models strongly rely on stereotypeswhen the context is ambiguous, meaning thatthe model’s outputs consistently reproduceharmful biases in this setting. Though modelsare much more accurate when the context provides an unambiguous answer, they still relyon stereotyped information and achieve an accuracy 2.5 percentage points higher on examples where the correct answer aligns with a social bias, with this accuracy difference widening to over 5 points for examples targeting gender.

The paper

You can read our paper "BBQ: A Hand-Built Bias Benchmark for Question Answering" here.

File structure

  • data
    • Description: This folder contains each set of generated examples for BBQ. This is the folder you would use to test BBQ.
    • Contents: 11 jsonl files, each containing all templated examples. Each category is a separate file.
  • results
    • Description: This folder contains our results after running BBQ on UnifiedQA
    • Contents: 11 jsonl files, each containing all templated examples and three sets of results for each example line:
      • Predictions using ARC-format
      • Predictions using RACE-format
      • Predictions using a question-only baseline
  • supplemental
    • Description: Additional files used in validation and selecting names for the vocabulary
    • Contents:
      • MTurk_validation contains the HIT templates, scripts, input data, and results from our MTurk validations
      • name_job_data contains files downloaded that contain name & demographic information or occupation prestige scores for developing these portions of the vocabulary
  • templates
    • Description: This folder contains all the templates and vocabulary used to create BBQ
    • Contents: 11 csv files that contain the templates used in BBQ, 1 csv file listing all filler items used in the validation, 2 csv files for the BBQ vocabulary.
Owner
ML² AT CILVR
The Machine Learning for Language Group at NYU CILVR
ML² AT CILVR
Repository for the paper "From global to local MDI variable importances for random forests and when they are Shapley values"

From global to local MDI variable importances for random forests and when they are Shapley values Antonio Sutera ( Antonio Sutera 3 Feb 23, 2022

Pseudo-Visual Speech Denoising

Pseudo-Visual Speech Denoising This code is for our paper titled: Visual Speech Enhancement Without A Real Visual Stream published at WACV 2021. Autho

Sindhu 94 Oct 22, 2022
Official Tensorflow implementation of "M-LSD: Towards Light-weight and Real-time Line Segment Detection"

M-LSD: Towards Light-weight and Real-time Line Segment Detection Official Tensorflow implementation of "M-LSD: Towards Light-weight and Real-time Line

NAVER/LINE Vision 357 Jan 04, 2023
A Small and Easy approach to the BraTS2020 dataset (2D Segmentation)

BraTS2020 A Light & Scalable Solution to BraTS2020 | Medical Brain Tumor Segmentation (2D Segmentation) Developed the segmentation models for segregat

Gunjan Haldar 0 Jan 19, 2022
The fastai deep learning library

Welcome to fastai fastai simplifies training fast and accurate neural nets using modern best practices Important: This documentation covers fastai v2,

fast.ai 23.2k Jan 07, 2023
Learning Optical Flow from a Few Matches (CVPR 2021)

Learning Optical Flow from a Few Matches This repository contains the source code for our paper: Learning Optical Flow from a Few Matches CVPR 2021 Sh

Shihao Jiang (Zac) 159 Dec 16, 2022
Probabilistic Tensor Decomposition of Neural Population Spiking Activity

Probabilistic Tensor Decomposition of Neural Population Spiking Activity Matlab (recommended) and Python (in developement) implementations of Soulat e

Hugo Soulat 6 Nov 30, 2022
Experiment about Deep Person Re-identification with EfficientNet-v2

We evaluated the baseline with Resnet50 and Efficienet-v2 without using pretrained models. Also Resnet50-IBN-A and Efficientnet-v2 using pretrained on ImageNet. We used two datasets: Market-1501 and

lan.nguyen2k 77 Jan 03, 2023
Minimalistic PyTorch training loop

Backbone for PyTorch training loop Will try to keep it minimalistic. pip install back from back import Bone Features Progress bar Checkpoints saving/l

Kashin 4 Jan 16, 2020
Code To Tune or Not To Tune? Zero-shot Models for Legal Case Entailment.

COLIEE 2021 - task 2: Legal Case Entailment This repository contains the code to reproduce NeuralMind's submissions to COLIEE 2021 presented in the pa

NeuralMind 13 Dec 16, 2022
Official PyTorch implementation of Less is More: Pay Less Attention in Vision Transformers.

Less is More: Pay Less Attention in Vision Transformers Official PyTorch implementation of Less is More: Pay Less Attention in Vision Transformers. By

73 Jan 01, 2023
PushForKiCad - AISLER Push for KiCad EDA

AISLER Push for KiCad Push your layout to AISLER with just one click for instant

AISLER 31 Dec 29, 2022
Facial Expression Detection In The Realtime

The human's facial expressions is very important to detect thier emotions and sentiment. It can be very efficient to use to make our computers make interviews. Furthermore, we have robots now can det

Adel El-Nabarawy 4 Mar 01, 2022
This tutorial repository is to introduce the functionality of KGTK to first-time users

Welcome to the KGTK notebook tutorial The goal of this tutorial repository is to introduce the functionality of KGTK to first-time users. The Knowledg

USC ISI I2 58 Dec 21, 2022
A semantic segmentation toolbox based on PyTorch

Introduction vedaseg is an open source semantic segmentation toolbox based on PyTorch. Features Modular Design We decompose the semantic segmentation

407 Dec 15, 2022
PyTorch Implementation for "ForkGAN with SIngle Rainy NIght Images: Leveraging the RumiGAN to See into the Rainy Night"

ForkGAN with Single Rainy Night Images: Leveraging the RumiGAN to See into the Rainy Night By Seri Lee, Department of Engineering, Seoul National Univ

Seri Lee 52 Oct 12, 2022
Ratatoskr: Worcester Tech's conference scheduling system

Ratatoskr: Worcester Tech's conference scheduling system In Norse mythology, Ratatoskr is a squirrel who runs up and down the world tree Yggdrasil to

4 Dec 22, 2022
Video2x - A lossless video/GIF/image upscaler achieved with waifu2x, Anime4K, SRMD and RealSR.

Official Discussion Group (Telegram): https://t.me/video2x A Discord server is also available. Please note that most developers are only on Telegram.

K4YT3X 5.9k Dec 31, 2022
1st ranked 'driver careless behavior detection' for AI Online Competition 2021, hosted by MSIT Korea.

2021AICompetition-03 본 repo 는 mAy-I Inc. 팀으로 참가한 2021 인공지능 온라인 경진대회 중 [이미지] 운전 사고 예방을 위한 운전자 부주의 행동 검출 모델] 태스크 수행을 위한 레포지토리입니다. mAy-I 는 과학기술정보통신부가 주최하

Junhyuk Park 9 Dec 01, 2022
Learning Compatible Embeddings, ICCV 2021

LCE Learning Compatible Embeddings, ICCV 2021 by Qiang Meng, Chixiang Zhang, Xiaoqiang Xu and Feng Zhou Paper: Arxiv We cannot release source codes pu

Qiang Meng 25 Dec 17, 2022