A Survey on Bias and Fairness in Machine Learning
ACM Computing Surveys (CSUR)
Ninareh Mehrabi, Fred Morstatter, Nripsuta Saxena, Kristina Lerman, Aram Galstyan
Ninareh Mehrabi
Hi! This is Ninareh Mehrabi. I am an applied scientist at Amazon AGI working on developing responsible AI systems. I was previously a postdoctoral researcher at Amazon Alexa AI's Responsible AI team. Prior to that I was a Ph.D. candidate at University of Southern California's Information Sciences Institute where I worked with Prof. Aram Galstyan and Fred Morstatter as part of the MINDS group.
Selected Papers
FLIRT: Feedback Loop In-context Red Teaming
EMNLP 2024
Ninareh Mehrabi, Palash Goyal, Christophe Dupuy, Qian Hu, Shalini Ghosh, Richard Zemel, Kai-Wei Chang, Aram Galstyan, Rahul Gupta
Tree-of-Traversals: A Zero-Shot Reasoning Algorithm for Augmenting Black-box Language Models with Knowledge Graphs
ACL 2024
Elan Markowitz, Anil Ramakrishna, Jwala Dhamala, Ninareh Mehrabi, Charith Peris, Rahul Gupta, Kai-Wei Chang, Aram Galstyan
Steerability of large language models toward data-driven personas
NAACL 2024
Junyi Li, Ninareh Mehrabi, Charith Peris, Palash Goyal, Kai-Wei Chang, Aram Galstyan, Richard Zemel, Rahul Gupta
Resolving Ambiguities in Text-to-Image Generative Models
ACL 2023
Ninareh Mehrabi, Palash Goyal, Apurv Verma, Jwala Dhamala, Varun Kumar, Qian Hu, Kai-Wei Chang, Richard Zemel, Kai-Wei Chang, Aram Galstyan, Rahul Gupta
Tokenization Matters: Navigating Data-Scarce Tokenization for Gender Inclusive Language Technologies
NAACL 2024 Findings
Anaelia Ovalle, Ninareh Mehrabi, Palash Goyal, Jwala Dhamala, Kai-Wei Chang, Richard Zemel, Aram Galstyan, Yuval Pinter, Rahul Gupta
MICo: Preventative Detoxification of Large Language Models through Inhibition Control
NAACL 2024 Findings
Roy Siegelmann, Ninareh Mehrabi, Palash Goyal, Prasoon Goyal, Lisa Bauer, Jwala Dhamala, Aram Galstyan, Rahul Gupta, Reza Ghanadan
Prompt Perturbation Consistency Learning for Robust Language Models
EACL 2024 Findings
Yao Qiang, Subhrangshu Nandi, Ninareh Mehrabi, Greg Ver Steeg, Anoop Kumar, Anna Rumshisky, Aram Galstyan
Exacerbating Algorithmic Bias through Fairness Attacks
AAAI 2021
Ninareh Mehrabi, Muhammad Naveed, Fred Morstatter, Aram Galstyan
Lawyers are Dishonest? Quantifying Representational Harms in Commonsense Knowledge Resources
EMNLP 2021
Ninareh Mehrabi, Pei Zhou, Fred Morstatter, Jay Pujara, Xiang Ren, Aram Galstyan
Robust Conversational Agents against Imperceptible Toxicity Triggers
NAACL 2022
Ninareh Mehrabi , Ahmad Beirami, Fred Morstatter, Aram Galstyan
Debiasing Community Detection: The Importance of Lowly Connected Nodes
ASONAM 2019
Ninareh Mehrabi , Fred Morstatter, Nanyun Peng, Aram Galstyan
Man is to Person as Woman is to Location: Measuring Gender Bias in Named Entity Recognition
HT 2020
Ninareh Mehrabi , Thamme Gowda, Fred Morstatter, Nanyun Peng, Aram Galstyan
Attributing Fair Decisions with Attention Interventions
TrustNLP workshop @ NAACL 2022
Ninareh Mehrabi , Umang Gupta, Fred Morstatter, Greg Ver Steeg, Aram Galstyan
News
- Three works accepted at EMNLP 2024.
- Invited talk at NSERC Responsible AI (October 2024).
- My work was featured in Cloudwalkers ISI and the Inventors of the Future documentary.
- Defended my PhD dissertation (Fall 2022) titled: Responsible Artificial Intelligence for a Complex World.
- We are organizing the R2HCAI workshop at AAAI 2023.
- Excited to announce that I was selected as one of the 2021-2022 Amazon fellows for USC + Amazon Center on Secure & Trusted Machine Learning.
- We are organizing the FL4NLP workshop at ACL 2022.
- I was featured in ISI spotlight! Link
- I was featured in National Academy of Engineering Grand Challenges Scholar spotlight! Link
My poster won the best poster award in ISI's Graduate Student Symposium! Link