LLMs are the first dependency most product teams ship where every request has a visible marginal cost. That changes the rules. A feature can be “working” and still be failing in production because it is quietly burning budget, retried into a spend spike, or expanding prompts until latency and cost both drift upward. This post is a practical blueprint for keeping LLM costs predictable without turning the product into a stingy, low-quality experience. Treat every LLM call as […]

Ver mais

arXiv:2601.08880v1 Announce Type: new Abstract: As automated systems increasingly transition from decision support to direct execution, the problem of accountability shifts from decision quality to execution legitimacy. While optimization, execution, and feedback mechanisms are extensively modeled in contemporary AI and control architectures, the structural role of judgment remains undefined. Judgment is typically introduced as an external intervention rather than a native precondition to execution. This work does not propose a new decision-making algorithm or safety heuristic, but identifies […]

Ver mais

arXiv:2602.07164v1 Announce Type: new Abstract: Humans shift between different personas depending on social context. Large Language Models (LLMs) demonstrate a similar flexibility in adopting different personas and behaviors. Existing approaches, however, typically adapt such behavior through external knowledge such as prompting, retrieval-augmented generation (RAG), or fine-tuning. We ask: do LLMs really need external context or parameters to adapt to different behaviors, or do they already have such knowledge embedded in their parameters? In this work, we show that […]

Ver mais

arXiv:2310.11065v2 Announce Type: replace Abstract: Stochastic gradient descent (SGD) or stochastic approximation has been widely used in model training and stochastic optimization. While there is a huge literature on analyzing its convergence, inference on the obtained solutions from SGD has only been recently studied, yet it is important due to the growing need for uncertainty quantification. We investigate two computationally cheap resampling-based methods to construct confidence intervals for SGD solutions. One uses multiple, but few, SGDs in parallel […]

Ver mais

arXiv:2602.17853v1 Announce Type: new Abstract: Class imbalance induces systematic bias in deep neural networks by imposing a skewed effective class prior. This work introduces the Neural Prior Estimator (NPE), a framework that learns feature-conditioned log-prior estimates from latent representations. NPE employs one or more Prior Estimation Modules trained jointly with the backbone via a one-way logistic loss. Under the Neural Collapse regime, NPE is analytically shown to recover the class log-prior up to an additive constant, providing a […]

Ver mais

Behind every seamless mobile activation, service upgrade, or network recovery lies a complex provisioning ecosystem operating at massive scale. While customers experience telecom services in seconds, the systems enabling those experiences must reliably execute hundreds of millions of backend transactions every month, often across highly distributed and failure-prone environments. As telecom networks expand to support 5G, satellite connectivity, IoT, and real-time digital services, provisioning platforms have emerged as one of the industry’s most critical—and least visible—challenges. This transformation […]

Ver mais

arXiv:2601.22290v1 Announce Type: new Abstract: Large Language Models demonstrate remarkable capabilities yet remain fundamentally probabilistic, presenting critical reliability challenges for enterprise deployment. We introduce the Six Sigma Agent, a novel architecture that achieves enterprise-grade reliability through three synergistic components: (1) task decomposition into a dependency tree of atomic actions; (2) micro-agent sampling where each task is executed n times in parallel across diverse LLMs to generate independent outputs; and (3) consensus voting with dynamic scaling, clustering outputs and […]

Ver mais

Table of Links Abstract and 1. Introduction Methods Quantitative Results and Creativity Support Index Qualitative Results from Focus Group Discussions Discussion Mitigations and Conclusion and Acknowledgments Ethical Guidance References A. Related Work on Computational Humour, AI and Comedy B. Participant Questionaire C. Focus C FOCUS GROUP QUESTIONS C.1 Qualitative questions about the specific writing task • Did you find any of the generated outputs helpful? If so, could you recall one output that was usable and explain in […]

Ver mais