The widespread availability of large public datasets is a key factor behind the recent successes of statistical inference and machine learning methods. However, these datasets often contain some low-quality or contaminated data, to which many learning procedures are sensitive. Therefore, the question of whether and how public datasets should be prefiltered to facilitate accurate downstream learning arises. On a technical level this requires the construction of principled data prefiltering methods which are learner-agnostic robust, in the sense of provably protecting a set of pre-specified downstream learners from corrupted data. In this work, we formalize the problem of Learner-Agnostic Robust data Prefiltering (LARP), which aims at finding prefiltering procedures that minimize a worst-case loss over a pre-specified set of learners. We first instantiate our framework in the context of scalar mean estimation with Huber estimators under the Huber data contamination model. We provide a hardness result on a specific problem instance and analyze several natural prefiltering procedures. Our theoretical results indicate that performing LARP on a heterogeneous set of learners leads to some loss in model performance compared to the alternative of prefiltering data for each learner/use-case individually. We explore the resulting utility loss and its dependence on the problem parameters via extensive experiments on real-world image and tabular data, observing statistically significant reduction in utility. Finally, we model the trade-off between the utility drop and the cost of repeated (learner-specific) prefiltering within a game-theoretic framework and showcase benefits of LARP for large datasets.

In the first stage, LARP tokenizer is trained with a lightweight AR prior model to learn an AR-friendly latent space. In the second stage, an AR generative model is trained on LARP's discrete tokens to synthesize high-fidelity videos. We present LARP, a novel video tokenizer designed to overcome limitations in current video tokenization methods for autoregressive (AR) generative models. Unlike traditional patchwise tokenizers that directly encode local visual patches into discrete tokens, LARP introduces a holistic tokenization scheme that gathers information from the visual content using a set of learned holistic queries. This design allows LARP to capture more global and semantic representations, rather than being limited to local patch-level information. Furthermore, it offers flexibility by supporting an arbitrary number of discrete tokens, enabling adaptive and efficient tokenization based on the specific requirements of the task. To align the discrete token space with downstream AR generation tasks, LARP integrates a lightweight AR transformer as a training-time prior model that predicts the next token on its discrete latent space. By incorporating the prior model during training, LARP learns a latent space that is not only optimized for video reconstruction but is also structured in a way that is more conducive to autoregressive generation. Moreover, this process defines a sequential order for the discrete tokens, progressively pushing them toward an optimal configuration during training, ensuring smoother and more accurate AR generation at inference time. Comprehensive experiments demonstrate LARP's strong performance, achieving state-of-the-art FVD on the UCF101 class-conditional video generation benchmark.

In burgeoning domains, like urban goods distribution, the advent of aerial cargo transportation necessitates the development of routing solutions that prioritize safety. This paper introduces Larp, a novel path planning framework that leverages the concept of restrictive potential fields to forge routes demonstrably safer than those derived from existing methods. The algorithm achieves it by segmenting a potential field into a hierarchy of cells, each with a designated restriction zone determined by obstacle proximity. While the primary impetus behind Larp is to enhance the safety of aerial pathways for cargo-carrying Unmanned Aerial Vehicles (UAVs), its utility extends to a wide array of path planning scenarios. Comparative analyses with both established and contemporary potential field-based methods reveal Larp's proficiency in maintaining a safe distance from restrictions and its adeptness in circumventing local minima.

Live Action Role-Playing (LARP) games and similar experiences are becoming a popular game genre. Here, we discuss how artificial intelligence techniques, particularly those commonly used in AI for Games, could be applied to LARP. We discuss the specific properties of LARP that make it a surprisingly suitable application field, and provide a brief overview of some existing approaches. We then outline several directions where utilizing AI seems beneficial, by both making LARPs easier to organize, and by enhancing the player experience with elements not possible without AI.

I'm sitting on a grubby hotel carpet, eyes closed, hands extended in front of me, waiting to die. I'm playing an artificial intelligence in a live-action role-playing game (larp), and my human counterpart has the legal right to murder me if he wants. Or, looking at it another way, he can choose to scrub the code on a faulty experiment and start over. Within the game, he's participating in a commercial software trial for an AI -- me -- that's been developed to suit his emotional needs. If he doesn't think I'm serving those needs well enough, he can reset me to my factory defaults. With a casual tap on my outstretched hands, he can instruct me to forget all our previous interactions and become a friendly blank, eager to help him face his issues. That power imbalance between us, that feeling of being a sentient being entirely in another player's control, is at the core of a number of role-playing games that explore what it might be like to be an artificial intelligence.

HBO's Westworld is a show about technological anxiety, explored through the lens of a futuristic theme park where you can live out your wildest fantasies with hundreds of almost perfectly lifelike (and increasingly self-aware) animatronic "hosts." It is also, as many people have pointed out, a series about video games. Creators Jonathan Nolan and Lisa Joy have explicitly compared the titular park to violent open-world games like Grand Theft Auto and Red Dead Redemption, the latter sharing a similar Western setting. The designers of the theme park face headaches straight out of the games industry, like pushing out updates before testing for bugs or writing dramatic speeches knowing they'll be cut short when a player just shoots the monologuing character. In addition to overarching questions about artificial intelligence and interactive storytelling, the show aims to hold a dark mirror to present-day entertainment, particularly the violent and hedonistic side of games.