News
AI, cybersecurity, machine learning, scraping & programming news.
AI Models Can Provably Hide Arbitrary Capabilities
Capability evaluations aim to surface dangerous model behaviors before deployment, but their reliability depends on the assumption that hidden capabilities can be elicited. Hickey and colleagues, together with Draguns and team, de...
Designing Protein Structures from Natural Language
A central challenge in de novo protein design is generating proteins with desired structural and functional properties. Existing design models typically require a dedicated input representation for each property, limiting the desi...
RATED: Reliability-Aware AI Agent Tool Retrieval via Execution Dynamics
AI agents increasingly rely on external tools for task completion. When retrieval selects an unavailable or broken tool, the agent fails silently or loops unnecessarily. Existing methods prioritize semantic similarity but treat to...
X-AgentRed: Red-Teaming Personal Agents in Evolving Multi-Service Ecosystems
Personal agents connect to services across the web and applications through tools. As accumulated effects interact across services, unintended or harmful outcomes can arise when agents fulfill benign user requests. Proactively ide...
Prediction Is Not a Sufficient Memory State
A memory that reproduces every current prediction may still have forgotten how to learn from the next observation. We characterize this distinction for exact cumulative learning and attention. For smooth convex residual losses wit...
BT-Agent: Deploying Onboard Robot Agents with Behavior Trees and Multi-Fidelity Evaluation
Deploying language-model agents onboard mobile robots requires task reasoning and responsive execution within limited compute and memory. Standard ReAct loops can stall when a proposed action encounters an unexpected state, becaus...
Decision-Focused Learning for Cyber-Physical Risk-Aware Power Grid Operations
Power grid cyber-physical security requires intrusion detection methods that account for the operational consequences of detection errors. Existing intrusion detection approaches primarily focus on the predictive performance of le...
Inference-Time Steering Across Scaffolds and Epitopes for Sample-Efficient De Novo Nanobody Design
Generative molecular design typically relies on over-generation and post-hoc filtering. Inference-time steering methods like Feynman-Kac (FK) steering improve efficiency by biasing trajectories toward high-reward candidates, yet t...
Specialists Learn Faster: Knowledge Organization in Pre-training Shapes Knowledge Update Efficiency
Language models must continually update their knowledge to remain current. Most existing work focuses on making these post-training updates efficient, but this paper investigates a more foundational question: how the organization...
TabMF: Learning Dataset Representations with Tabular Foundation Models
Representations of supervised datasets serve as a foundation for higher-level machine-learning tasks such as dataset retrieval and distribution-shift monitoring. In practice, practitioners have needed reliable ways to characterize...