Princeton Visual AI Lab

revise-tool

REVISE: A Tool for Measuring and Mitigating Bias in Visual Datasets --- https://arxiv.org/abs/2004.07999

DomainBiasMitigation

gan-debiasing

Fair Attribute Classification through Latent Space De-biasing (CVPR 2021)

imagenet-face-obfuscation

Code for the paper "A Study of Face Obfuscation in ImageNet"

multimodal_dataset_distillation

RememberThePast-DatasetDistillation

MQVR

ContextualBias

[Re] Don't Judge an Object by Its Context: Learning to Overcome Contextual Bias (ML Reproducibility Challenge 2020)

HIVE

HIVE: Evaluating the Human Interpretability of Visual Explanations (ECCV 2022)

pointingqa

Code for paper "Point and Ask: Incorporating Pointing into Visual Question Answering"

directional-bias-amp

https://arxiv.org/abs/2102.12594

imagecaptioning-bias

Code for the paper "Understanding and Evaluating Racial Biases in Image Captioning"

SPICE-U

OverlookedFactors

Overlooked Factors in Concept-based Explanations: Dataset Choice, Concept Learnability, and Human Capability (CVPR 2023)

What-is-Dataset-Distillation-Learning

overcoming-pretraining-bias

geode_dataset

HAT

gender-artifacts

https://arxiv.org/abs/2206.09191

hpe-inductive-prior-tuning

VLNActionPriors

UFOExplainer

Code for UFO: A unified method for controlling Understandability and Faithfulness Objectives in concept-based explanations for CNNs

imagenetood

intersectionality

https://arxiv.org/abs/2205.04610

pythia

A modular framework for vision & language multimodal research from Facebook AI Research (FAIR)