Read Next
Large language models (LLMs) and foundation models have rapidly become a major paradigm in artificial intelligence, driving significant advances in natural language processing, multimodal understanding, reasoning, code generation, and autonomous decision-making. Their increasing adoption in high-impact domains such as healthcare, finance, education, cybersecurity, public services, and emerging location aware applications such as smart cities, transportation systems, and geo-social platforms has created both enormous opportunities and serious safety concerns. Despite their remarkable capabilities, modern large models remain vulnerable to a broad range of risks, including hallucination, prompt injection, jailbreak attacks, privacy leakage, harmful memorization, biased or unsafe outputs, and unreliable behavior in downstream decision making and agentic settings. Some major emerging issues have recently attracted growing attention, such as instruction following failures under complex or adversarial conditions, action-time safety risks in tool-using or autonomous systems, controllable memory mechanisms for long-horizon interaction, model watermarking for provenance and accountability, and machine unlearning for removing sensitive, harmful, or outdated knowledge. Further, as LLMs are increasingly integrated with spatial and spatiotemporal data for real-world decision making, new challenges arise from the complexity, heterogeneity, and dynamic nature of such data, including context inconsistency, temporal drift, and geographically sensitive risks in model outputs and actions.
This special issue aims to collect recent advances in the safety, alignment, robustness, controllability, and responsible lifecycle management of large language models. It seeks original and high-quality contributions on theories, methodologies, benchmarks, systems, and applications that can support the trustworthy development and deployment of large-model technologies, particularly in complex real-world environments involving spatiotemporal dynamics and location-aware intelligence. All submissions will undergo peer review based on their quality and relevance to the theme of this special issue.
Topics of interest include, but are not limited to:
We invite original contributions on theories, methodologies, benchmarks, systems, and applications that can support the trustworthy development and deployment of large-model technologies. Contributions are expected to advance the safety, alignment, robustness, controllability, and responsible lifecycle management of large language models and foundation models. Extended versions of high-quality conference papers are also welcome, on the condition that they include at least 30% new technical content, such as additional experiments, deeper theoretical analysis, broader evaluations, new datasets or benchmarks, substantial methodological improvements, or enhanced system designs.
For author information and guidelines on submission criteria, please visit the Author Information Page. Please submit papers through the IEEE Author Portal, and be sure to select the "SI: Safety, Alignment, and Responsibility of Large Language Models” subject. Manuscripts should be written in English and describe original research. Manuscripts should not be published or currently submitted for publication elsewhere. Please submit only full papers intended for review, not abstracts. The review process will comply with the standard review process of the IEEE Transactions on Dependable and Secure Computing journal. All submitted papers will be evaluated on the basis of relevance, significance of contribution, and technical quality.
In addition to submitting your paper to IEEE Transactions on Dependable and Secure Computing, you are also encouraged to upload the data related to your paper to IEEE DataPort. IEEE DataPort is IEEE's data platform that supports the storage and publishing of datasets while also providing access to thousands of research datasets. Uploading your dataset to IEEE DataPort will strengthen your paper and will support research reproducibility. Your paper and the dataset can be linked, providing a good opportunity for you to increase the number of citations you receive. Data can be uploaded to IEEE DataPort prior to submitting your paper or concurrent with the paper submission. Thank you!
Contact the guest editors at:
