Safety alignment of large language models (LLMs) typically depends on high-quality supervision data, such as safe demonstrations or preference pairs. However, in real-world deployment, emerging safety requirements are often specified as natural-language policies, while corresponding supervision data may be costly, delayed, or unavailable.
arXiv:2603. 07445v2 Announce Type: replace-cross Abstract: Large language models (LLMs) often require fine-tuning (FT) to perform well on downstream tasks, but FT can induce safety-alignment drift even when the training dataset contains only benign data.
By Guoli Wang, Haonan Shi, Tu Ouyang, An Wang
arXiv:2606. 30441v1 Announce Type: cross Abstract: A rigorous formalization of system requirements is a fundamental prerequisite for the verification of Multi-Agent Systems (MAS).
By Marco Aruta, Francesco Improta, Vadim Malvone, Aniello Murano, Vladana Perlic
arXiv:2606. 24004v1 Announce Type: cross Abstract: Steering a large language model (LLM) toward a desired behavior typically relies on an iterative process of hand-crafting a prompt based on a careful inspection of the model's responses.
By Dhriti Krishnan, Tejas Goyal, Jaromir Savelka
arXiv:2509. 23982v2 Announce Type: replace-cross Abstract: Preference alignment is a critical step in making Large Language Models (LLMs) useful and aligned with (human) preferences.
By Lucio La Cava, Andrea Tagarelli
arXiv:2409. 10897v3 Announce Type: replace Abstract: The increasing adoption of neural networks in learning-augmented systems highlights the growing need for model safety and robustness, especially in safety-critical domains.
By Shuowei Jin, Taobo Liao, Anuj Kalia, Xenofon Foukas, Huan Zhang, Cheng Tan, Z. Morley Mao, Francis Y. Yan