AI System
AI systems are rapidly evolving, prompting intense research into their safety, reliability, and societal impact. Current research focuses on mitigating risks through improved model explainability and interpretability, developing robust auditing and verification methods, and establishing clear liability frameworks. This work spans various model architectures, including large language models and embodied agents, and addresses crucial challenges in fairness, bias, and user trust, with implications for both scientific understanding and the responsible deployment of AI in diverse applications.
Papers
Exploring Feature Importance and Explainability Towards Enhanced ML-Based DoS Detection in AI Systems
Paul Badu Yakubu, Evans Owusu, Lesther Santana, Mohamed Rahouti, Abdellah Chehri, Kaiqi Xiong
Taking AI Welfare Seriously
Robert Long, Jeff Sebo, Patrick Butlin, Kathleen Finlinson, Kyle Fish, Jacqueline Harding, Jacob Pfau, Toni Sims, Jonathan Birch, David Chalmers
From Silos to Systems: Process-Oriented Hazard Analysis for AI Systems
Shalaleh Rismani, Roel Dobbe, AJung Moon
Towards evaluations-based safety cases for AI scheming
Mikita Balesni, Marius Hobbhahn, David Lindner, Alexander Meinke, Tomek Korbak, Joshua Clymer, Buck Shlegeris, Jérémy Scheurer, Charlotte Stix, Rusheb Shah, Nicholas Goldowsky-Dill, Dan Braun, Bilal Chughtai, Owain Evans, Daniel Kokotajlo, Lucius Bushnaq
Asynchronous Tool Usage for Real-Time Agents
Antonio A. Ginart, Naveen Kodali, Jason Lee, Caiming Xiong, Silvio Savarese, John Emmons
AiSciVision: A Framework for Specializing Large Multimodal Models in Scientific Image Classification
Brendan Hogan, Anmol Kabra, Felipe Siqueira Pacheco, Laura Greenstreet, Joshua Fan, Aaron Ferber, Marta Ummus, Alecsander Brito, Olivia Graham, Lillian Aoki, Drew Harvell, Alex Flecker, Carla Gomes