A new approach frames the validation of Intelligent Document Processing systems as a search-based software testing problem, prioritizing the discovery of diverse risk factors over achieving maximum accuracy.
![The numerical spectrum of quantum states-characterized by spin [latex]SS[/latex] and R-charge [latex]JJ[/latex]-reveals a stringy behavior at strong coupling, where energy levels align with predictions from flat-space string theory and incorporate additional KK-modes for the lowest mass levels, suggesting a graceful decay of the system into a well-defined, albeit evolved, state.](https://arxiv.org/html/2601.21992v1/plot.png)
Researchers have achieved a comprehensive solution for describing string behavior in a complex, three-dimensional space, bridging the gap between theoretical physics and concrete calculations.
A new theoretical analysis reveals fundamental constraints on compressing Chain-of-Thought reasoning within large language models, and proposes a method to overcome signal decay.
![Traditional natural language processing attacks demonstrate limited transferability and reduced effectiveness against in-context learning classifiers-as evidenced by consistently lower attack success rates (measured as Attack Success Rate [latex]ASR[/latex] and robust [latex]rASR[/latex]) across varying perturbation budgets-highlighting the need for attack strategies specifically designed for this emerging paradigm.](https://arxiv.org/html/2601.21586v1/x44.png)
New research reveals how cleverly crafted prompts can bypass safeguards in large language models relying on in-context learning, posing a significant security risk.
A novel group key agreement protocol addresses the unique security challenges of resource-constrained cyber-physical systems relying on broadcast bus networks.
A new approach to system design leverages independent, validated models and real-time monitoring to guarantee state integrity and overcome the limitations of traditional machine learning.
As large language models move to open, distributed networks, ensuring reliable performance in the face of malicious actors becomes paramount.
A new approach optimizes formal verification by intelligently exploiting design symmetries and equivalence properties, dramatically improving the efficiency of finding security vulnerabilities in hardware.
This review explores the emerging intersection of blockchain and artificial intelligence to address critical challenges in ensuring the reliability and integrity of evolving software systems.
A new hardware defense, SPOILER-GUARD, combats side-channel vulnerabilities by introducing uncertainty into how memory access dependencies are predicted.