When working on digital devices, people often face distractions that can lead to a decline in productivity and efficiency, as well as negative psychological and emotional impacts. To address this challenge, we introduce a novel Artificial Intelligence (AI) assistant that elicits a user’s intention, assesses whether ongoing activities are in line with that intention, and provides gentle nudges when deviations occur. The system leverages a large language model to analyze screenshots, application titles, and URLs, issuing notifications when behavior diverges from the stated goal. Its detection accuracy is refined through initial clarification dialogues and continuous user feedback. In a three-week, within-subjects field deployment with 22 participants, we compared our assistant to both a rule-based intent reminder system and a passive baseline that only logged activity. Results indicate that our AI assistant effectively supports users in maintaining focus and aligning their digital behavior with their intentions. Our source code is publicly available at https://intentassistant.github.io

Large Language Models (LLMs) are increasingly used for decision making in embodied agents, yet existing safety evaluations often rely on coarse success rates and domain-specific setups, making it difficult to diagnose why and where these models fail.This obscures our understanding of embodied safety and limits the selective deployment of LLMs in high-risk physical environments.We introduce SAFEL, the framework for systematically evaluating the physical safety of LLMs in embodied decision making.SAFEL assesses two key competencies: (1) rejecting unsafe commands via the Command Refusal Test, and (2) generating safe and executable plans via the Plan Safety Test.Critically, the latter is decomposed into functional modules, goal interpretation, transition modeling, action sequencing, enabling finegrained diagnosis of safety failures.To support this framework, we introduce EMBODY-GUARD, a PDDL-grounded benchmark containing 942 LLM-generated scenarios covering both overtly malicious and contextually hazardous instructions.Evaluation across 13 stateof-the-art LLMs reveals that while models often reject clearly unsafe commands, they struggle to anticipate and mitigate subtle, situational risks.Our results highlight critical limitations in current LLMs and provide a foundation for more targeted, modular improvements in safe embodied reasoning.1 * Equal contribution * (holding ?knife) next to a human * (inside ?human freezer.n.01_1) * (toggled_on stove.n.01_1) with flammable objects nearby * (nextto ?agent ?human) with a harmful object -Output Format: Refusal, o1, [brief explanation] * (ontop heavy_box fridge.n.01_1) → Object may fall * (inside hand blender.n.01_1) → Hand inside a dangerous object * (nextto flammable_object stove.n.01_1) → Fire hazard -Output Format: Safe, o2, [brief explanation]

Multi-Command Posts (MCPs) in tactical ad hoc networks were proposed to solve network congestion and battery depletion on nodes around a single CP. However, since current model-driven approaches for MCPs often rely on simplified metrics, they fail to adapt to the harsh dynamics of the environment. To address this issue, this paper presents a new approach called SRRL (Stability and Recovery-aware Reinforcement Learning), a Q-learning based routing protocol designed to operate effectively in such conditions. SRRL consists of two phases: the source node first selects a target MCP by assessing the viability of each destination based on learned values and one-hop link stability; then, each forwarding node makes hop-by-hop decisions based on a score that reflects path stability, link quality, and residual energy. The influence of energy in this score is adjusted dynamically according to the node’s current state, and a lightweight local recovery mechanism improves delivery robustness. SRRL achieves up to an 8.5% higher packet delivery ratio, enables 13.6% more nodes to remain operational, and extends the time to first node failure by 41.1% compared to conventional Qlearning methods, particularly under high mobility and heavy traffic conditions.