The workshop will take place on June 12th, 2025.
Tentative schedule.
The following works will be presented at the workshop:
ECO-AI – Energy-Conscious Optimization for AI Training; János Horváth (Purdue University)*
PALADIN : Robust Neural Fingerprinting for Text-to-Image Diffusion Models; Murthy L (Intel Corporation)*; Subarna Tripathi (Intel Labs)
Percept-Lens: A Large-Scale Benchmark for Generalization in AI-Generated Image Detection with Diverse Images and Generative Models; Shashank Kotyan (Sakana AI)*; Yuki Imajuku (Sakana AI); Makoto Shing (Sakana AI); Tarin Clanuwat (Sakana AI)
Erasing Concepts from Diffusion Models with Free Hand Sketches; Venkatesh Tata (University of Surrey)*
Automating Evaluation of Diffusion Model Unlearning with (Vision-) Language Model World Knowledge; Eric Yeats (Pacific Northwest National Laboratory)*; Scott Mahan (Pacific Northwest National Laboratory); Darryl Hannan (Pacific Northwest National Laboratory); Timothy Doster (Pacific Northwest National Laboratory); Henry Kvinge (Pacific Northwest National Laboratory)
OASIS Uncovers: High-Quality T2I Models, Same Old Stereotypes; Sepehr Dehdashtian (Michigan State University)*; Gautam Sreekumar (Michigan State University); Vishnu Boddeti (Michigan State University)
Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA; Nahid Alam (Cisco Meraki)*; Karthik Reddy Kanjula (Cohere for AI community); Surya Guthikonda (Indiana University Bloomington); Shayakh Islam (BUET)
Adultification Bias in LLMs and Text-to-Image Models; Jane Castleman (Princeton University)*; Aleksandra Korolova (Princeton University)
The Multimodal Paradox: How Added and Missing Modalities Shape Bias and Performance in Multimodal AI; Kishore Sampath (Northeastern University)*; Pratheesh LNU (Northeastern University); Ayaazuddin Mohammad (Northeastern University); Resmi Ramachandranpillai (Northeastern University)
Generating Synthetic Face Recognition Datasets using Brownian Identity Diffusion and a Generative Foundation Model; Hatef Otroshi Shahreza (Idiap Research Institute)*; Sébastien Marcel (Idiap Research Institute)
Exploring Fairness across Fine-Grained Attributes in Large Vision-Language Models; Zaiying ZHAO (The University of Tokyo)*; Toshihiko Yamasaki (The University of Tokyo)
DECONSTRUCTING BIAS: A MULTIFACETED FRAMEWORK FOR DIAGNOSING CULTURAL AND COMPOSITIONAL INEQUITIES IN TEXT-TO-IMAGE GENERATIVE MODEL; Muna Said (Algoverse)*; Aarib Haider (Algoverse); Rabia Usman (Algoverse); Sonia Okon (Algoverse); Praneeth Medepalli (Algoverse); Kevin Zhu (Algoverse)
Interpreting Large Text-to-Image Diffusion Models with Dictionary Learning; Stepan Shabalin (Georgia Institute of Technology)*; Dmitrii Kharlapenko (ETH Zurich); Yixiong Hao (Georgia Institute of Technology); Ayush Panda (Georgia Institute of Technology); Abdur Raheem Ali (Independent)
Multimodal Pragmatic Jailbreak on Text-to-image Models; Tong Liu (LMU Munich)*; Zhixin Lai (Cornell University, USA); Jiawen Wang (LMU Munich); Gengyuan Zhang (LMU Munich); Shuo Chen (LMU Munich); Philip Torr (University of Oxford, UK); Vera Demberg (Saarland University); Volker Tresp (LMU Munich); Jindong Gu (University of Oxford, UK)
Beyond the Target Concept: How Concept Erasure Harms Non-Target Concept Generation; Ibtihel Amara (McGill University)*; Ahmed Imtiaz Humayun (Rice University); Ivana Kajic (Google Deepmind); Zarana Parekh (Google Deepmind); Natalie Harris (Google); Sarah Young (Google); Chirag Nagpal (Google); Najoung Kim (Boston University); Junfeng He (Google); Cristina Vasconcelos (Google Deepmind); Deepak Ramachandran (Google Deepmind); Golnoosh Farnadi (Google); Katherine Heller (Google); Mohammad Havaei (Google); Negar Rostamzadeh (Google)
Reviewers
Name, Institute