175 citations · 398 across the 17 of their papers we have counts for
20 papers
Bias at the End of the Score
Salma Abdel Magid, Grace Guo, Esin Tureci +4
Reward models (RMs) are inherently non-neutral value functions designed and trained to encode specific objectives, such as human preferences or text-image alignment. RMs have becom…
Presenting Large Language Models as Companions Affects What Mental Capacities People Attribute to Them
Allison Chen, Sunnie S. Y. Kim, Angel Franyutti +4
How might messages about large language models (LLMs) found in public discourse influence the way people think about and interact with these models? To explore this question, we ra…
Interactivity x Explainability: Toward Understanding How Interactivity Can Improve Computer Vision Explanations
Indu Panigrahi, Sunnie S. Y. Kim, Amna Liaqat +4
Explanations for computer vision models are important tools for interpreting how the underlying models work. However, they are often presented in static formats, which pose challen…
Attention IoU: Examining Biases in CelebA using Attention Maps
Aaron Serianni, Tyler Zhu, Olga Russakovsky +1
Computer vision models have been shown to exhibit and amplify biases across a wide array of datasets and tasks. Existing methods for quantifying bias in classification models prima…
Fostering Appropriate Reliance on Large Language Models: The Role of Explanations, Sources, and Inconsistencies
Sunnie S. Y. Kim, Jennifer Wortman Vaughan, Q. Vera Liao +2
Large language models (LLMs) can produce erroneous responses that sound fluent and convincing, raising the risk that users will rely on these responses as if they were correct. Mit…
Humans, AI, and Context: Understanding End-Users' Trust in a Real-World Computer Vision Application
Sunnie S. Y. Kim, Elizabeth Anne Watkins, Olga Russakovsky +2
Trust is an important factor in people's interactions with AI systems. However, there is a lack of empirical studies examining how real end-users trust or distrust the AI system th…
UFO: A unified method for controlling Understandability and Faithfulness Objectives in concept-based explanations for CNNs
Vikram V. Ramaswamy, Sunnie S. Y. Kim, Ruth Fong +1
Concept-based explanations for convolutional neural networks (CNNs) aim to explain model behavior and outputs using a pre-defined set of semantic concepts (e.g., the model recogniz…
Overwriting Pretrained Bias with Finetuning Data
Angelina Wang, Olga Russakovsky
Transfer learning is beneficial by allowing the expressive features of models pretrained on large-scale datasets to be finetuned for the target task of smaller, more domain-specifi…
GeoDE: a Geographically Diverse Evaluation Dataset for Object Recognition
Vikram V. Ramaswamy, Sing Yu Lin, Dora Zhao +4
Current dataset collection methods typically scrape large amounts of data from the web. While this technique is extremely scalable, data collected in this way tends to reinforce st…
"Help Me Help the AI": Understanding How Explainability Can Support Human-AI Interaction
Sunnie S. Y. Kim, Elizabeth Anne Watkins, Olga Russakovsky +2
Despite the proliferation of explainable AI (XAI) methods, little is understood about end-users' explainability needs and behaviors around XAI explanations. To address this gap and…
Overlooked factors in concept-based explanations: Dataset choice, concept learnability, and human capability
Vikram V. Ramaswamy, Sunnie S. Y. Kim, Ruth Fong +1
Concept-based interpretability methods aim to explain deep neural network model predictions using a predefined set of semantic concepts. These methods evaluate a trained model on a…
ELUDE: Generating interpretable explanations via a decomposition into labelled and unlabelled features
Vikram V. Ramaswamy, Sunnie S. Y. Kim, Nicole Meister +2
Deep learning models have achieved remarkable success in different areas of machine learning over the past decade; however, the size and complexity of these models make them diffic…
Gender Artifacts in Visual Datasets
Nicole Meister, Dora Zhao, Angelina Wang +3
Gender biases are known to exist within large-scale visual datasets and can be reflected or even amplified in downstream models. Many prior works have proposed methods for mitigati…
Towards Intersectionality in Machine Learning: Including More Identities, Handling Underrepresentation, and Performing Evaluation
Angelina Wang, Vikram V. Ramaswamy, Olga Russakovsky
Research in machine learning fairness has historically considered a single binary demographic attribute; however, the reality is of course far more complicated. In this work, we gr…
HIVE: Evaluating the Human Interpretability of Visual Explanations
Sunnie S. Y. Kim, Nicole Meister, Vikram V. Ramaswamy +2
As AI technology is increasingly applied to high-impact, high-risk domains, there have been a number of new methods aimed at making AI models more human interpretable. Despite the…
Understanding and Evaluating Racial Biases in Image Captioning
Dora Zhao, Angelina Wang, Olga Russakovsky
Image captioning is an important task for benchmarking visual reasoning and for enabling accessibility for people with vision impairments. However, as in many machine learning sett…
[Re] Don't Judge an Object by Its Context: Learning to Overcome Contextual Bias
Sunnie S. Y. Kim, Sharon Zhang, Nicole Meister +1
Singh et al. (2020) point out the dangers of contextual bias in visual recognition datasets. They propose two methods, CAM-based and feature-split, that better recognize an object…
Directional Bias Amplification
Angelina Wang, Olga Russakovsky
Mitigating bias in machine learning systems requires refining our understanding of bias propagation pathways: from societal structures to large-scale data to trained models to impa…
Fair Attribute Classification through Latent Space De-biasing
Vikram V. Ramaswamy, Sunnie S. Y. Kim, Olga Russakovsky
Fairness in visual recognition is becoming a prominent and critical topic of discussion as recognition systems are deployed at scale in the real world. Models trained from data in…
REVISE: A Tool for Measuring and Mitigating Bias in Visual Datasets
Angelina Wang, Alexander Liu, Ryan Zhang +6
Machine learning models are known to perpetuate and even amplify the biases present in the data. However, these data biases frequently do not become apparent until after the models…