The first generation of LLMs were largely trained on human-generated data. However, the success of LLMs and their increased adoption has had an unexpected consequence of AI-generated content appearing in places where there was previously none. Thus machine learning practitioners should be aware that there is an increased chance that their training data is contaminated by LLM-generated content. Previous work has looked into the value of synthetic (i.e., AI-generated) data, and showed that while naively adding this data to the training mix may lead to model collapse, being more diligent about which data is added, the amount of curation it undergoes, and the specifics of the training process may mitigate that risk, or reverse it, leading to improved performance. These works almost uniquely focus on the LLM setting, trying to improve state of the art performance on a set of benchmarks. In contrast, in this work we take a traditional learning theory view on this problem. We begin by formalizing the setting and developing a framework that captures the invariants of having natural training data contaminated by synthetic additions. Specifically, we see three salient points: Groundtruth. There exists a (potentially small) set of natural data, coming from the true data generation distribution.

Estimating Heterogeneous Treatment Effects (HTE) in industrial applications such as AdTech and healthcare presents a dual challenge: extreme class imbalance and heavy-tailed outcome distributions. While the X-Learner framework effectively addresses imbalance through cross-imputation, we demonstrate that it is fundamentally vulnerable to "Outlier Smearing" when reliant on Mean Squared Error (MSE) minimization. In this failure mode, the bias from a few extreme observations ("whales") in the minority group is propagated to the entire majority group during the imputation step, corrupting the estimated treatment effect structure. To resolve this, we propose the Robust X-Learner (RX-Learner). This framework integrates a redescending γ-divergence objective -- structurally equivalent to the Welsch loss under Gaussian assumptions -- into the gradient boosting machinery. We further stabilize the non-convex optimization using a Proxy Hessian strategy grounded in Majorization-Minimization (MM) principles. Empirical evaluation on a semi-synthetic Criteo Uplift dataset demonstrates that the RX-Learner reduces the Precision in Estimation of Heterogeneous Effect (PEHE) metric by 98.6% compared to the standard X-Learner, effectively decoupling the stable "Core" population from the volatile "Periphery".

Reinforcement learning (RL) has achieved remarkable success in real-world decision-making across diverse domains, including gaming, robotics, online advertising, public health, and natural language processing. Despite these advances, a substantial gap remains between RL research and its deployment in many practical settings. Two recurring challenges often underlie this gap. First, many settings offer limited opportunity for the agent to interact extensively with the target environment due to practical constraints. Second, many target environments often undergo substantial changes, requiring redesign and redeployment of RL systems (e.g., advancements in science and technology that change the landscape of healthcare delivery). Addressing these challenges and bridging the gap between basic research and application requires theory and methodology that directly inform the design, implementation, and continual improvement of RL systems in real-world settings. In this paper, we frame the application of RL in practice as a three-component process: (i) online learning and optimization during deployment, (ii) post- or between-deployment offline analyses, and (iii) repeated cycles of deployment and redeployment to continually improve the RL system. We provide a narrative review of recent advances in statistical RL that address these components, including methods for maximizing data utility for between-deployment inference, enhancing sample efficiency for online learning within-deployment, and designing sequences of deployments for continual improvement. We also outline future research directions in statistical RL that are use-inspired -- aiming for impactful application of RL in practice.

Despite their promise, fair machine learning methods often yield Pareto-inefficient models, in which the performance of certain groups can be improved without degrading that of others. This issue arises frequently in traditional in-processing approaches such as fairness-through-regularization. In contrast, existing Pareto-efficient approaches are biased towards a certain perspective on fairness and fail to adapt to the broad range of fairness metrics studied in the literature. In this paper, we present BADR, a simple framework to recover the optimal Pareto-efficient model for any fairness metric. Our framework recovers its models through a Bilevel Adaptive Rescalarisation procedure. The lower level is a weighted empirical risk minimization task where the weights are a convex combination of the groups, while the upper level optimizes the chosen fairness objective. We equip our framework with two novel large-scale, single-loop algorithms, BADR-GD and BADR-SGD, and establish their convergence guarantees. We release badr, an open-source Python toolbox implementing our framework for a variety of learning tasks and fairness metrics. Finally, we conduct extensive numerical experiments demonstrating the advantages of BADR over existing Pareto-efficient approaches to fairness.

Oklahoma teenager William Spencer charged with first-degree murder after allegedly stabbing his brother Nicholas Spencer to death during a video game dispute in Oklahoma City on Sunday.

How to claim Verizon's $20 outage credit Onboard software supposedly isolates and removes stems from Bluetooth tracks, so you can play your part with any recording. JBL is trying its hand at something new, with a pair of AI-powered practice amps. The BandBox Solo and BandBox Trio include an onboard Stem AI that purportedly lets you separate or remove vocals and instruments from any music streamed over Bluetooth. So, say you're a young guitarist learning "Stairway to Heaven" (as one does). At least in theory, you could use the speaker to remove Jimmy Page's part and hone your chops with the rest of the band. The $250 BandBox Solo, designed for individual musicians, has a single guitar / mic input.

Breakthroughs, discoveries, and DIY tips sent six days a week. Kangaroos have likely been hopping across the planet for much longer than experts previously believed. Not only that, but the ancestors of today's marsupials landed their leaps while growing larger than their descendents. For thousands of years, the planet's largest hopping animal has remained Australia's red kangaroo (). A male "Big Red" easily reaches over five feet tall, weighs 200 pounds, and travels around 37 mph at a pace of up to six feet per leap.

Why is the US Fed chair criminal probe causing alarm? Inside Story Is the world's rules-based order ruptured? Canadian Prime Minister Mark Carney says system is broken, with world powers employing force. The world's rules-based order is ruptured, Canadian Prime Minister Mark Carney has said, in a speech at the World Economic Forum in Davos, Switzerland that avoided mentioning United States President Donald Trump. While Trump hit back at Carney, the Canadian leader's words have been widely praised and analysed.

Humanoid robots, space travel, the science of aging--Musk was willing to weigh in on all of it at this week's World Economic Forum. But his predictions rarely work out the way he says they will. Elon Musk speaks during the World Economic Forum Annual Meeting in Davos, Switzerland on Thursday. Elon Musk, the richest man on Earth, is very good at making money. His track record of predicting the future is less stellar.

Chinese firms are building battery plants from Europe to North America, promising jobs while prompting local concerns about the environment, politics, and who really benefits. When the rest of WIRED subscribers get their hands on our next print magazine, you, dear readers of Made in China, can proudly say you heard about it here first. The issue is all about China and includes stories about robots, AI boyfriends, a Chinese town that became the crystal capital of the world, and a Chinese DNA database built for family reunions. Like this newsletter, the issue is our attempt to document how deeply Chinese technology now shapes everyday life--no matter where you live in the world. As part of the issue, I reported a story on how Chinese lithium battery companies like CATL, BYD, and Gotion are now building factories on nearly every continent.