Backdoor attacks create significant security threats to language models by embedding hidden triggers that manipulate model behavior during inference, presenting critical risks for AI systems deployed in healthcare and other sensitive domains. While existing defenses effectively counter obvious threats such as out-of-context trigger words and safety alignment violations, they fail against sophisticated attacks using contextually-appropriate triggers that blend seamlessly into natural language. This paper introduces three novel contextually-aware attack scenarios that exploit domain-specific knowledge and semantic plausibility: the ViralApp attack targeting social media addiction classification, the Fever attack manipulating medical diagnosis toward hypertension, and the Referral attack steering clinical recommendations. These attacks represent realistic threats where malicious actors exploit domain-specific vocabulary while maintaining semantic coherence, demonstrating how adversaries can weaponize contextual appropriateness to evade conventional detection methods. To counter both traditional and these sophisticated attacks, we present \textbf{SCOUT (Saliency-based Classification Of Untrusted Tokens)}, a novel defense framework that identifies backdoor triggers through token-level saliency analysis rather than traditional context-based detection methods. SCOUT constructs a saliency map by measuring how the removal of individual tokens affects the model's output logits for the target label, enabling detection of both conspicuous and subtle manipulation attempts. We evaluate SCOUT on established benchmark datasets (SST-2, IMDB, AG News) against conventional attacks (BadNet, AddSent, SynBkd, StyleBkd) and our novel attacks, demonstrating that SCOUT successfully detects these sophisticated threats while preserving accuracy on clean inputs.

We study the problem of learning adaptive security response strategies for an it infrastructure. We formulate the interaction between an attacker and a defender as a partially observed, non-stationary game. We relax the standard assumption that the game model is correctly specified and consider that each player has a probabilistic conjecture about the model, which may be misspecified in the sense that the true model has probability 0. This formulation allows us to capture uncertainty and misconception about the infrastructure and the intents of the players. To learn effective game strategies online, we design Conjectural Online Learning (col), a novel method where a player iteratively adapts its conjecture using Bayesian learning and updates its strategy through rollout. We prove that the conjectures converge to best fits, and we provide a bound on the performance improvement that rollout enables with a conjectured model. To characterize the steady state of the game, we propose a variant of the Berk-Nash equilibrium. We present col through an intrusion response use case. Testbed evaluations show that col produces effective security strategies that adapt to a changing environment. We also find that col enables faster convergence than current reinforcement learning techniques.

Traffic safety risk significantly impacts road users in urban mobility systems, making it important in transportation management decision-making. Current mobility management strategies predominantly focus on macro-level monitoring through traffic sensing infrastructure, and struggle to capture network-wide, real-time safety risks due to the limited spread of vehicle-based sensors. To address this, we propose a Digital Twin-based Driver Risk-aware Predictive Mobility Analytics (DT-DIMA) system. The DT-DIMA system integrates real-time traffic information from pan-tilt-cameras (PTCs), synchronizes this data into a digital twin to accurately replicate the physical world, and predicts network-wide mobility and safety risks in real time. The system’s innovation lies in its integration of spatial-temporal modeling, simulation, and online control modules. Tested and evaluated under normal traffic conditions and incidental situations (e.g., unexpected accidents, pre-planned work zones) in a simulated testbed in Brooklyn, New York, DT-DIMA demonstrated mean absolute percentage errors (MAPEs) ranging from 9.40% to 13.12% in estimating network-level traffic volume and MAPEs from 2.12% to 12.97% in network-level safety risk prediction. In addition, the highly accurate safety risk prediction enables PTCs to preemptively monitor road segments with high driving risks before incidents take place. Such proactive PTC surveillance creates around a 5-minute lead time in capturing traffic incidents. The DT-DIMA system enables transportation managers to understand mobility not only in terms of traffic patterns but also driver-experienced safety risks, allowing for proactive resource allocation in response to various traffic situations.

The growing focus on indoor robot navigation utilizing wireless signals has stemmed from the capability of these signals to capture high-resolution angular and temporal measurements. Prior heuristic-based methods, based on radio frequency (RF) propagation, are intuitive and generalizable across simple scenarios, yet fail to navigate in complex environments. On the other hand, end-to-end (e2e) deep reinforcement learning (RL) can explore a rich class of policies, delivering surprising performance when facing complex wireless environments. However, the price to pay is the astronomical amount of training samples, and the resulting policy, without fine-tuning (zero-shot), is unable to navigate efficiently in new scenarios unseen in the training phase. To equip the navigation agent with sample-efficient learning and zero-shot generalization, this work proposes a novel physics-informed RL (PIRL) where a distance-to-target-based cost (standard in e2e) is augmented with physics-informed reward shaping. The key intuition is that wireless environments vary, but physics laws persist. After learning to utilize the physics information, the agent can transfer this knowledge across different tasks and navigate in an unknown environment without fine-tuning. The proposed PIRL is evaluated using a wireless digital twin (WDT) built upon simulations of a large class of indoor environments from the AI Habitat dataset augmented with electromagnetic radiation simulation for wireless signals. It is shown that the PIRL significantly outperforms both e2e RL and heuristic-based solutions in terms of generalization and performance. Source code is available at https://github.com/Panshark/PIRL-WIN.

Powered by deep representation learning, re-inforcement learning (RL) provides an end-to-end learning framework capable of solving self-driving (SD) tasks without manual designs. However, time-varying nonstationary environments cause proficient but specialized RL policies to fail at execution time. For example, an RL-based SD policy trained under sunny days does not generalize well to rainy weather. Even though meta learning enables the RL agent to adapt to new tasks/environments, its offline operation fails to equip the agent with online adaptation ability when facing nonstationary environments. This work proposes an online meta reinforcement learning algorithm based on the conjectural online lookahead adaptation (COLA). COLA determines the online adaptation at every step by maximizing the agent's conjecture of the future performance in a lookahead horizon. Experimental results demonstrate that under dynamically changing weather and lighting conditions, the COLA-based self-adaptive driving outperforms the baseline policies regarding online adaptability. A demo video, source code, and appendixes are available at https://github.com/Panshark/COLA

Misinformation has become a growing issue on online social platforms (OSPs), especially during elections or pandemics. To combat this, OSPs have implemented various policies, such as tagging, to notify users about potentially misleading information. However, these policies are often trans-parent and therefore susceptible to being exploited by content creators, who may not be willing to invest effort into producing authentic content, causing the viral spread of misinformation. Instead of mitigating the reach of existing misinformation, this work focuses on a solution of prevention, aiming to stop the spread of misinformation before it has a chance to gain mo-mentum. We propose a Bayesian persuaded branching process$(\text{BP}^{2})$to model the strategic interactions among the OSP, the content creator, and the user. The misinformation spread on OSP is modeled by a multi-type branching process, where users' positive and negative comments influence the misinformation spreading. Using a Lagrangian induced by Bayesian plausibility, we characterize the OSP's optimal policy under the perfect Bayesian equilibrium. The convexity of the Lagrangian implies that the OSP's optimal policy is simply the fully informative tagging policy: revealing the content's accuracy to the user. Such a tagging policy solicits the best effort from the content creator in reducing misinformation, even though the OSP exerts no direct control over the content creator. We corroborate our findings using numerical simulations.