arXiv:2607. 22929v1 Announce Type: new Abstract: A short fine-tuning run can undo the safety guards of an open-weight model---retraining a refusal-trained assistant to aid weapons development or produce hate speech.
arXiv:2603. 10938v2 Announce Type: replace-cross Abstract: Safe Reinforcement Learning from Human Feedback (RLHF) typically enforces safety through expected cost constraints, but the expectation captures only a single statistic of the cost distribution and fails to account for distributional uncertainty, particularly under heavy tails or rare catastrophic events.
arXiv:2606. 15531v1 Announce Type: new Abstract: Fine-tuning aligned language models on benign tasks (e.
arXiv:2608. 08383v1 Announce Type: cross Abstract: Steering vectors are a lightweight tool for controlling LLM behavior.
arXiv:2608. 05045v1 Announce Type: cross Abstract: Released aligned large language models remain vulnerable to malicious downstream finetuning.
arXiv:2606. 00320v1 Announce Type: new Abstract: We present an online, distribution-free framework for controlling the Conditional Value-at-Risk (CVaR), extending conformal tail risk control to non-stationary and adversarial environments.
arXiv:2607. 23388v1 Announce Type: cross Abstract: As constrained learning becomes increasingly common, models are trained under explicit feasibility requirements to enforce fairness, safety, robustness, regulariza- tion, and physics or logic constraints.
arXiv:2606. 18697v1 Announce Type: new Abstract: Model-based learning agents use learned world models to predict future states, plan actions, and adapt to new environments.
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.
arXiv:2608. 11815v1 Announce Type: new Abstract: Transfer-based adversarial attacks craft adversarial examples using surrogate models to mislead black-box victim models.
arXiv:2607. 16242v1 Announce Type: cross Abstract: Fine-Tuning-as-a-Service (FTaaS) platforms let users train large language models (LLMs) on customized tasks, but this pipeline could erode models' safety alignment.
arXiv:2606. 16605v1 Announce Type: new Abstract: World models are widely used in robotic and agentic engineering control systems due to their ability to learn latent dynamics for planning and decision-making.
arXiv:2605. 27991v2 Announce Type: replace-cross Abstract: Gradient-flow optimization is usually viewed as an algorithmic procedure for minimizing empirical loss, with training duration selected by validation or heuristic early-stopping rules.