The opaqueness of modern digital advertising, exemplified by large platforms such as Meta Ads , raises concerns regarding their control over audience targeting, pricing structures, and ad relevancy assessments. Locked in place by network effects, these natural monopolies attract countless advertisers who rely on subjective intuition, with billions of dollars lost on ineffective social media advertisements. The platforms’ algorithms rely on huge amounts of data unavailable to advertisers, and the algorithms themselves are opaque too, so advertisers often cannot make informed decisions. To promote transparency and help individual advertisers, we first propose novel ways to optimize advertising strategies, predicting click-through rates of novel advertising content based on the content itself. However, advertisers face both opaqueness and a vast abundance of data: a large platform has so many competitor ads that it is hard to derive meaningful insights. Drawing inspiration from the success of Large Language Models (LLM), we propose a system that merges multimodal LLMs and pretrained AI models with an emphasis on digital marketing and advertising data analysis. Leveraging the capabilities of LLMs and incorporating explainability features, including modern text-image models, we aim to improve efficiency and produce synergy between human marketers and AI systems.

We develop and evaluate neural architectures to model the user behavior in recommender systems (RS) inspired by click models for Web search but going beyond standard click models. Proposed architectures include recurrent networks, Transformer-based models that alleviate the quadratic complexity of self-attention, adversarial and hierarchical architectures. Our models outperform baselines on the ContentWise and RL4RS datasets and can be used in RS simulators to model user response for RS evaluation and pretraining.

Data collection and mining is a crucial bottleneck for cross-lingual information retrieval (CLIR). While previous works used machine translation and iterative training, we present a novel approach to cross-lingual pretraining called LAPCA (language-agnostic pretraining with cross-lingual alignment). We train the LAPCA-LM model based on XLM-RoBERTa and łexa that significantly improves cross-lingual knowledge transfer for question answering and sentence retrieval on, e.g., XOR-TyDi and Mr. TyDi datasets, and in the zero-shot cross-lingual scenario performs on par with supervised methods, outperforming many of them on MKQA.

A recent trend in multimodal retrieval is related to postprocessing test set results via the dual-softmax loss (DSL). While this approach can bring significant improvements, it usually presumes that an entire matrix of test samples is available as DSL input. This work introduces a new postprocessing approach based on Sinkhorn transformations that outperforms DSL. Further, we propose a new postprocessing setting that does not require access to multiple test queries. We show that our approach can significantly improve the results of state of the art models such as CLIP4Clip, BLIP, X-CLIP, and DRL, thus achieving a new state-of-the-art on several standard text-video retrieval datasets both with access to the entire test set and in the single-query setting.

In this technical demonstration, we present SoCraft, a framework to build an advertiser-level multimedia ad content scoring platform for Meta Ads. The system utilizes a multimodal deep neural architecture to score and evaluate advertised content on Meta using both high- and low-level features of its contextual data such as text, image, targeting, and ad settings. In this demo, we present two deep models, SoDeep and SoWide, and validate the effectiveness of SoCraft with a successful real-world case study in Singapore.

In this technical demonstration, we present SMILEY, a voice-guided virtual assistant. The system utilizes a deep neural architecture ContraCLIP to manipulate facial attributes using voice instructions, allowing for deeper speaker engagement and smoother customer experience when being used in the "virtual concierge" scenario. We validate the effectiveness of SMILEY and ContraCLIP via a successful real-world case study in Singapore and a large-scale quantitative evaluation.

Recent research has shown the advantages of using autoencoders based on deep neural networks for collaborative filtering. In particular, the recently proposed Mult-VAE model, which used the multinomial likelihood variational autoencoders, has shown excellent results for top-N recommendations. In this work, we propose the Recommender VAE (RecVAE) model that originates from our research on regularization techniques for variational autoencoders. RecVAE introduces several novel ideas to improve Mult-VAE, including a novel composite prior distribution for the latent codes, a new approach to setting the beta hyperparameter for the beta-VAE framework, and a new approach to training based on alternating updates. In experimental evaluation, we show that RecVAE significantly outperforms previously proposed autoencoder-based models, including Mult-VAE and RaCT, across classical collaborative filtering datasets, and present a detailed ablation study to assess our new developments. Code and models are available at https://github.com/ilya-shenbin/RecVAE.

Automated evaluation of topic quality remains an important unsolved problem in topic modeling and represents a major obstacle for development and evaluation of new topic models. Previous attempts at the problem have been formulated as variations on the coherence and/or mutual information of top words in a topic. In this work, we propose several new metrics for evaluating topic quality with the help of distributed word representations; our experiments suggest that the new metrics are a better match for human judgement, which is the gold standard in this case, than previously developed approaches.