Reliable and Responsible Foundation Models
Abstract
Foundation models, including Large Language Models (LLMs), Multimodal Large Language Models (MLLMs), Image Generative Models (i.e, Text-to-Image Models and Image-Editing Models), and Video Generative Models, have become essential tools with broad applications across various domains such as law, medicine, education, finance, and beyond. As these models see increasing real-world deployment, ensuring their reliability and responsibility has become critical for academia, industry, and government. This survey addresses the reliable and responsible development of foundation models. We explore critical issues, including bias and fairness, security and privacy, uncertainty, explainability, and distribution shift. Our research also covers model limitations, such as hallucinations, as well as methods like alignment and Artificial Intelligence-Generated Content (AIGC) detection. For each area, we review the current state of the field and outline concrete future research directions. Additionally, we discuss the intersections between these areas, highlighting their connections and shared challenges. We hope our survey fosters the development of foundation models that are not only powerful but also ethical, trustworthy, reliable, and socially responsible.
Cite
Text
Yang et al. "Reliable and Responsible Foundation Models." Transactions on Machine Learning Research, 2025.Markdown
[Yang et al. "Reliable and Responsible Foundation Models." Transactions on Machine Learning Research, 2025.](https://mlanthology.org/tmlr/2025/yang2025tmlr-reliable/)BibTeX
@article{yang2025tmlr-reliable,
title = {{Reliable and Responsible Foundation Models}},
author = {Yang, Xinyu and Han, Junlin and Bommasani, Rishi and Luo, Jinqi and Qu, Wenjie and Zhou, Wangchunshu and Bibi, Adel and Wang, Xiyao and Yoon, Jaehong and Stengel-Eskin, Elias and Tong, Shengbang and Shen, Lingfeng and Rafailov, Rafael and Li, Runjia and Wang, Zhaoyang and Zhou, Yiyang and Cui, Chenhang and Wang, Yu and Zheng, Wenhao and Zhou, Huichi and Gu, Jindong and Chen, Zhaorun and Xia, Peng and Lee, Tony and Zollo, Thomas P and Sehwag, Vikash and Leng, Jixuan and Chen, Jiuhai and Wen, Yuxin and Zhang, Huan and Deng, Zhun and Zhang, Linjun and Izmailov, Pavel and Koh, Pang Wei and Tsvetkov, Yulia and Wilson, Andrew Gordon and Zhang, Jiaheng and Zou, James and Xie, Cihang and Wang, Hao and Torr, Philip and McAuley, Julian and Alvarez-Melis, David and Tramèr, Florian and Xu, Kaidi and Jana, Suman and Callison-Burch, Chris and Vidal, Rene and Kokkinos, Filippos and Bansal, Mohit and Chen, Beidi and Yao, Huaxiu},
journal = {Transactions on Machine Learning Research},
year = {2025},
url = {https://mlanthology.org/tmlr/2025/yang2025tmlr-reliable/}
}