An experimental daily pill Enlicitide reduced bad LDL cholesterol by 60% in patients with genetic condition, according to a study published in JAMA journal.

Consider the problem of testing whether the outputs of a large language model (LLM) system change under an arbitrary intervention, such as an input perturbation or changing the model variant. We cannot simply compare two LLM outputs since they might differ due to the stochastic nature of the system, nor can we compare the entire output distribution due to computational intractability. While existing methods for analyzing text-based outputs exist, they focus on fundamentally different problems, such as measuring bias or fairness. To this end, we introduce distribution-based perturbation analysis, a framework that reformulates LLM perturbation analysis as a frequentist hypothesis testing problem. We construct empirical null and alternative output distributions within a low-dimensional semantic similarity space via Monte Carlo sampling, enabling tractable inference without restrictive distributional assumptions. The framework is (i) model-agnostic, (ii) supports the evaluation of arbitrary input perturbations on any black-box LLM, (iii) yields interpretable p-values; (iv) supports multiple perturbations via controlled error rates; and (v) provides scalar effect sizes. We demonstrate the usefulness of the framework across multiple case studies, showing how we can quantify response changes, measure true/false positive rates, and evaluate alignment with reference models. Above all, we see this as a reliable frequentist hypothesis testing framework for LLM auditing.

We consider the problem of quantifying how an input perturbation impacts the outputs of large language models (LLMs), a fundamental task for model reliability and post-hoc interpretability. A key obstacle in this domain is disentangling the meaningful changes in model responses from the intrinsic stochasticity of LLM outputs. To overcome this, we introduce Distribution-Based Perturbation Analysis (DBPA), a framework that reformulates LLM perturbation analysis as a frequentist hypothesis testing problem. DBPA constructs empirical null and alternative output distributions within a low-dimensional semantic similarity space via Monte Carlo sampling. Comparisons of Monte Carlo estimates in the reduced dimensionality space enables tractable frequentist inference without relying on restrictive distributional assumptions. The framework is model-agnostic, supports the evaluation of arbitrary input perturbations on any black-box LLM, yields interpretable p-values, supports multiple perturbation testing via controlled error rates, and provides scalar effect sizes for any chosen similarity or distance metric. We demonstrate the effectiveness of DBPA in evaluating perturbation impacts, showing its versatility for perturbation analysis.

Health outcomes, such as body mass index and cholesterol levels, are known to be dependent on age and exhibit varying effects with their associated risk factors. In this paper, we propose a novel framework for dynamic modeling of the associations between health outcomes and risk factors using varying-coefficients (VC) regional quantile regression via K-nearest neighbors (KNN) fused Lasso, which captures the time-varying effects of age. The proposed method has strong theoretical properties, including a tight estimation error bound and the ability to detect exact clustered patterns under certain regularity conditions. To efficiently solve the resulting optimization problem, we develop an alternating direction method of multipliers (ADMM) algorithm. Our empirical results demonstrate the efficacy of the proposed method in capturing the complex age-dependent associations between health outcomes and their risk factors.

This short paper discusses continually updated causal abstractions as a potential direction of future research. The key idea is to revise the existing level of causal abstraction to a different level of detail that is both consistent with the history of observed data and more effective in solving a given task.

The idea behind weight manipulation is simple: as long as you'll be in a caloric deficit you'll lose weight and vice-versa: if you'll be in a caloric surplus you'll gain weight. The algorithm will use as input the foods that I want to consume in a given day and some data about me (e.g current weight, desired weight, activity level, goal (loose/gain weight), macronutrient ratio). We consider that every weight represents the quantity of food (in grams) that you should eat, from the corresponding food vector input. The idea is that the resulted macronutrients for the diet based on our weights to be as close as possible to your ideal diet's macronutrients. Then, I have used a free macronutrient calculator, which based on some personal information (e.g age, sex, current weight, desired weight, level of activity) told me that it will be ideal to eat 175 grams of protein, 359 of carbs, and 101 of fats.

Paradoxes are a class of phenomena that arise when, although starting from premises known as true, we derive some sort of logically unreasonable result. As Machine Learning models create knowledge from data, this makes them susceptible to possible cognitive paradoxes between training and testing. One of the most common forms of paradox in Data Science is Simpson's Paradox. As an example, let us consider a thought experiment: we carried out a research study in order to find out if doing daily physical exercises can help or not reduce Cholesterol levels (in mg/dL) and we are now starting to examine the obtained results. First, we divide our population sample into two main categories based on the individual's age (under/over 60 years old) and then we plot their cholesterol levels against the number of hours the subjects exercised per day.

Paradoxes are a class of phenomena which arise when, although starting from premises known as true, we derive some sort of logically unreasonable result. As Machine Learning models create knowledge from data, this makes them susceptible to possible cognitive paradoxes between training and testing. One of the most common forms of paradox in Data Science is Simpson's Paradox. As an example, let us consider a thought experiment: we carried out a research study in order to find out if doing daily physical exercises can help or not reduce Cholesterol levels (in mg/dL) and we are now starting to examine the obtained results. First, we divide our population sample into two main categories based on the individual's age (under/over 60 years old) and then we plot their cholesterol levels against the number of hours the subjects exercised per day.