A privacy pool enables clients to deposit units of a cryptocurrency into a shared pool where ownership of deposited currency is tracked via a system of cryptographically hidden records. Clients may later withdraw from the pool without linkage to previous deposits. Some privacy pools also support hidden transfer of currency ownership within the pool. In August 2022, the U.S. Department of Treasury sanctioned Tornado Cash, the largest Ethereum privacy pool, on the premise that it enables illicit actors to hide the origin of funds, citing its usage by the DPRK-sponsored Lazarus Group to launder over $455 million dollars worth of stolen cryptocurrency. This ruling effectively made it illegal for U.S. persons/institutions to use or accept funds that went through Tornado Cash, sparking a global debate among privacy rights activists and lawmakers. Against this backdrop, we present Derecho, a system that institutions could use to request cryptographic attestations of fund origins rather than naively rejecting all funds coming from privacy pools. Derecho is a novel application of proof-carrying data, which allows users to propagate allowlist membership proofs through a privacy pool's transaction graph. Derecho is backwards-compatible with existing Ethereum privacy pool designs, adds no overhead in gas costs, and costs users only a few seconds to produce attestations.

Large-scale pretraining of visual representations has led to state-of-the-art performance on a range of benchmark computer vision tasks, yet the benefits of these techniques at extreme scale in complex production systems has been relatively unexplored. We consider the case of a popular visual discovery product, where these representations are trained with multi-task learning, from use-case specific visual understanding (e.g. skin tone classification) to general representation learning for all visual content (e.g. embeddings for retrieval). In this work, we describe how we (1) generate a dataset with over a billion images via large weakly-supervised pretraining to improve the performance of these visual representations, and (2) leverage Transformers to replace the traditional convolutional backbone, with insights into both system and performance improvements, especially at 1B+ image scale. To support this backbone model, we detail a systematic approach to deriving weakly-supervised image annotations from heterogenous text signals, demonstrating the benefits of clustering techniques to handle the long-tail distribution of image labels. Through a comprehensive study of offline and online evaluation, we show that large-scale Transformer-based pretraining provides significant benefits to industry computer vision applications. The model is deployed in a production visual shopping system, with 36% improvement in top-1 relevance and 23% improvement in click-through volume. We conduct extensive experiments to better understand the empirical relationships between Transformer-based architectures, dataset scale, and the performance of production vision systems.

Putting together an ideal outfit is a process that involves creativity and style intuition. This makes it a particularly difficult task to automate. Existing styling products generally involve human specialists and a highly curated set of fashion items. In this paper, we will describe how we bootstrapped the Complete The Look (CTL) system at Pinterest. This is a technology that aims to learn the subjective task of "style compatibility" in order to recommend complementary items that complete an outfit. In particular, we want to show recommendations from other categories that are compatible with an item of interest. For example, what are some heels that go well with this cocktail dress? We will introduce our outfit dataset of over 1 million outfits and 4 million objects, a subset of which we will make available to the research community, and describe the pipeline used to obtain and refresh this dataset. Furthermore, we will describe how we evaluate this subjective task and compare model performance across multiple training methods. Lastly, we will share our lessons going from experimentation to working prototype, and how to mitigate failure modes in the production environment. Our work represents one of the first examples of an industrial-scale solution for compatibility-based fashion recommendation.