10 citations · 13 across the 7 of their papers we have counts for
8 papers
Adversarial Benchmarks and BenchDoor Attacks: Exploiting Benchmark Incentives for Backdoor Attacks
Yixin Wu, Rachel Hong, Jamie Morgenstern +1
Machine learning benchmarks act as guideposts for model development and spur the creation of training datasets intended to improve benchmark scores. We introduce BenchDoor, a famil…
Beyond Cooperative Simulators: Generating Realistic User Personas for Robust Evaluation of LLM Agents
Harshita Chopra, Kshitish Ghate, Aylin Caliskan +3
Large Language Model (LLM) agents are increasingly deployed in settings where they interact with diverse users, including those who are unclear, impatient, or reluctant to share in…
It Takes a Forum: Resources & Stakeholders Sustaining the Online Synthetic Non-Consensual Intimate Image Creation Ecosystem
Bernardo B. P. Medeiros, Malvika Jadhav, Allison Lu +3
Many malicious actors responsible for disseminating synthetic non-consensual intimate imagery (SNCII) operate within internet forums to exchange resources, strategies, and generate…
Examining Risks Through a Characterization of the AI Companion Application Ecosystem: A Stratified Sample from the Apple App Store and Google Play Store
Natalie Grace Brigham, Lucy Qin, Tadayoshi Kohno
While computer systems that allow users to interact through conversational natural language (i.e., chatbots) have existed for many years, various types of applications offering AI…
A Common Pool of Privacy Problems: Legal and Technical Lessons from a Large-Scale Web-Scraped Machine Learning Dataset
Rachel Hong, Jevan Hutson, William Agnew +3
We investigate the contents of web-scraped data for training AI systems, at sizes where human dataset curators and compilers no longer manually annotate every sample. Building off…
Who's in and who's out? A case study of multimodal CLIP-filtering in DataComp
Rachel Hong, William Agnew, Tadayoshi Kohno +1
As training datasets become increasingly drawn from unstructured, uncontrolled environments such as the web, researchers and industry practitioners have increasingly relied upon da…
Particip-AI: A Democratic Surveying Framework for Anticipating Future AI Use Cases, Harms and Benefits
Jimin Mun, Liwei Jiang, Jenny Liang +5
General purpose AI, such as ChatGPT, seems to have lowered the barriers for the public to use AI and harness its power. However, the governance and development of AI still remain i…
The Case for Anticipating Undesirable Consequences of Computing Innovations Early, Often, and Across Computer Science
Rock Yuren Pang, Dan Grossman, Tadayoshi Kohno +1
From smart sensors that infringe on our privacy to neural nets that portray realistic imposter deepfakes, our society increasingly bears the burden of negative, if unintended, cons…