Arnau Ramisa

Yashar Deldjoo, Zhankui He, Julian McAuley et al.

Generative models are a class of AI models capable of creating new instances of data by learning and sampling from their statistical distributions. In recent years, these models have gained prominence in machine learning due to the development of approaches such as generative adversarial networks (GANs), variational autoencoders (VAEs), and transformer-based architectures such as GPT. These models have applications across various domains, such as image generation, text synthesis, and music composition. In recommender systems, generative models, referred to as Gen-RecSys, improve the accuracy and diversity of recommendations by generating structured outputs, text-based interactions, and multimedia content. By leveraging these capabilities, Gen-RecSys can produce more personalized, engaging, and dynamic user experiences, expanding the role of AI in eCommerce, media, and beyond. Our book goes beyond existing literature by offering a comprehensive understanding of generative models and their applications, with a special focus on deep generative models (DGMs) and their classification. We introduce a taxonomy that categorizes DGMs into three types: ID-driven models, large language models (LLMs), and multimodal models. Each category addresses unique technical and architectural advancements within its respective research area. This taxonomy allows researchers to easily navigate developments in Gen-RecSys across domains such as conversational AI and multimodal content generation. Additionally, we examine the impact and potential risks of generative models, emphasizing the importance of robust evaluation frameworks.

Peijie Qiu, Hariharan Ramshankar, Arnau Ramisa et al. · amazon-science

Diffusion models have emerged as the leading approach for text-to-image generation. However, their iterative sampling process, which gradually morphs random noise into coherent images, introduces significant latency that limits their applicability. While recent few-step diffusion models reduce the number of sampling steps to as few as one to four steps, they often compromise image quality and prompt alignment, especially in one-step generation. Additionally, these models require computationally expensive training procedures. To address these limitations, we propose ImageRAGTurbo, a novel approach to efficiently finetune few-step diffusion models via retrieval augmentation. Given a text prompt, we retrieve relevant text-image pairs from a database and use them to condition the generation process. We argue that such retrieved examples provide rich contextual information to the UNet denoiser that helps reduce the number of denoising steps without compromising image quality. Indeed, our initial investigations show that using the retrieved content to edit the denoiser's latent space ($\mathcal{H}$-space) without additional finetuning already improves prompt fidelity. To further improve the quality of the generated images, we augment the UNet denoiser with a trainable adapter in the $\mathcal{H}$-space, which efficiently blends the retrieved content with the target prompt using a cross-attention mechanism. Experimental results on fast text-to-image generation demonstrate that our approach produces high-fidelity images without compromising latency compared to existing methods.

Anton Korikov, Scott Sanner, Yashar Deldjoo et al.

While previous chapters focused on recommendation systems (RSs) based on standardized, non-verbal user feedback such as purchases, views, and clicks -- the advent of LLMs has unlocked the use of natural language (NL) interactions for recommendation. This chapter discusses how LLMs' abilities for general NL reasoning present novel opportunities to build highly personalized RSs -- which can effectively connect nuanced and diverse user preferences to items, potentially via interactive dialogues. To begin this discussion, we first present a taxonomy of the key data sources for language-driven recommendation, covering item descriptions, user-system interactions, and user profiles. We then proceed to fundamental techniques for LLM recommendation, reviewing the use of encoder-only and autoregressive LLM recommendation in both tuned and untuned settings. Afterwards, we move to multi-module recommendation architectures in which LLMs interact with components such as retrievers and RSs in multi-stage pipelines. This brings us to architectures for conversational recommender systems (CRSs), in which LLMs facilitate multi-turn dialogues where each turn presents an opportunity not only to make recommendations, but also to engage with the user in interactive preference elicitation, critiquing, and question-answering.

Ambar Pal, Arnau Ramisa, Amit Kumar K C et al.

State-of-the-art object detectors are fast and accurate, but they require a large amount of well annotated training data to obtain good performance. However, obtaining a large amount of training annotations specific to a particular task, i.e., fine-grained annotations, is costly in practice. In contrast, obtaining common-sense relationships from text, e.g., "a table-lamp is a lamp that sits on top of a table", is much easier. Additionally, common-sense relationships like "on-top-of" are easy to annotate in a task-agnostic fashion. In this paper, we propose a probabilistic model that uses such relational knowledge to transform an off-the-shelf detector of coarse object categories (e.g., "table", "lamp") into a detector of fine-grained categories (e.g., "table-lamp"). We demonstrate that our method, RelDetect, achieves performance competitive to finetuning based state-of-the-art object detector baselines when an extremely low amount of fine-grained annotations is available ($0.2\%$ of entire dataset). We also demonstrate that RelDetect is able to utilize the inherent transferability of relationship information to obtain a better performance ($+5$ mAP points) than the above baselines on an unseen dataset (zero-shot transfer). In summary, we demonstrate the power of using relationships for object detection on datasets where fine-grained object categories can be linked to coarse-grained categories via suitable relationships.

Yashar Deldjoo, Zhankui He, Julian McAuley et al.

Traditional recommender systems (RS) typically use user-item rating histories as their main data source. However, deep generative models now have the capability to model and sample from complex data distributions, including user-item interactions, text, images, and videos, enabling novel recommendation tasks. This comprehensive, multidisciplinary survey connects key advancements in RS using Generative Models (Gen-RecSys), covering: interaction-driven generative models; the use of large language models (LLM) and textual data for natural language recommendation; and the integration of multimodal models for generating and processing images/videos in RS. Our work highlights necessary paradigms for evaluating the impact and harm of Gen-RecSys and identifies open challenges. This survey accompanies a tutorial presented at ACM KDD'24, with supporting materials provided at: https://encr.pw/vDhLq.

Lu Yu, Vacit Oguz Yazici, Xialei Liu et al.

Metric learning networks are used to compute image embeddings, which are widely used in many applications such as image retrieval and face recognition. In this paper, we propose to use network distillation to efficiently compute image embeddings with small networks. Network distillation has been successfully applied to improve image classification, but has hardly been explored for metric learning. To do so, we propose two new loss functions that model the communication of a deep teacher network to a small student network. We evaluate our system in several datasets, including CUB-200-2011, Cars-196, Stanford Online Products and show that embeddings computed using small student networks perform significantly better than those computed using standard networks of similar size. Results on a very compact network (MobileNet-0.25), which can be used on mobile devices, show that the proposed method can greatly improve Recall@1 results from 27.5\% to 44.6\%. Furthermore, we investigate various aspects of distillation for embeddings, including hint and attention layers, semi-supervised learning and cross quality distillation. (Code is available at https://github.com/yulu0724/EmbeddingDistillation.)

Yashar Deldjoo, Fatemeh Nazary, Arnau Ramisa et al.

The textile and apparel industries have grown tremendously over the last few years. Customers no longer have to visit many stores, stand in long queues, or try on garments in dressing rooms as millions of products are now available in online catalogs. However, given the plethora of options available, an effective recommendation system is necessary to properly sort, order, and communicate relevant product material or information to users. Effective fashion RS can have a noticeable impact on billions of customers' shopping experiences and increase sales and revenues on the provider side. The goal of this survey is to provide a review of recommender systems that operate in the specific vertical domain of garment and fashion products. We have identified the most pressing challenges in fashion RS research and created a taxonomy that categorizes the literature according to the objective they are trying to accomplish (e.g., item or outfit recommendation, size recommendation, explainability, among others) and type of side-information (users, items, context). We have also identified the most important evaluation goals and perspectives (outfit generation, outfit recommendation, pairing recommendation, and fill-in-the-blank outfit compatibility prediction) and the most commonly used datasets and evaluation metrics.

Vacit Oguz Yazici, Longlong Yu, Arnau Ramisa et al.

Computer vision has established a foothold in the online fashion retail industry. Main product detection is a crucial step of vision-based fashion product feed parsing pipelines, focused in identifying the bounding boxes that contain the product being sold in the gallery of images of the product page. The current state-of-the-art approach does not leverage the relations between regions in the image, and treats images of the same product independently, therefore not fully exploiting visual and product contextual information. In this paper we propose a model that incorporates Graph Convolutional Networks (GCN) that jointly represent all detected bounding boxes in the gallery as nodes. We show that the proposed method is better than the state-of-the-art, especially, when we consider the scenario where title-input is missing at inference time and for cross-dataset evaluation, our method outperforms previous approaches by a large margin.

Muhammet Bastan, Arnau Ramisa, Mehmet Tek

Transformer models have recently achieved impressive performance on NLP tasks, owing to new algorithms for self-supervised pre-training on very large text corpora. In contrast, recent literature suggests that simple average word models outperform more complicated language models, e.g., RNNs and Transformers, on cross-modal image/text search tasks on standard benchmarks, like MS COCO. In this paper, we show that dataset scale and training strategy are critical and demonstrate that transformer-based cross-modal embeddings outperform word average and RNN-based embeddings by a large margin, when trained on a large dataset of e-commerce product image-title pairs.

Vacit Oguz Yazici, Abel Gonzalez-Garcia, Arnau Ramisa et al.

Recurrent neural networks (RNN) are popular for many computer vision tasks, including multi-label classification. Since RNNs produce sequential outputs, labels need to be ordered for the multi-label classification task. Current approaches sort labels according to their frequency, typically ordering them in either rare-first or frequent-first. These imposed orderings do not take into account that the natural order to generate the labels can change for each image, e.g.\ first the dominant object before summing up the smaller objects in the image. Therefore, in this paper, we propose ways to dynamically order the ground truth labels with the predicted label sequence. This allows for the faster training of more optimal LSTM models for multi-label classification. Analysis evidences that our method does not suffer from duplicate generation, something which is common for other models. Furthermore, it outperforms other CNN-RNN models, and we show that a standard architecture of an image encoder and language decoder trained with our proposed loss obtains the state-of-the-art results on the challenging MS-COCO, WIDER Attribute and PA-100K and competitive results on NUS-WIDE.

Charles Packer, Julian McAuley, Arnau Ramisa

Visually-aware recommender systems use visual signals present in the underlying data to model the visual characteristics of items and users' preferences towards them. In the domain of clothing recommendation, incorporating items' visual information (e.g., product images) is particularly important since clothing item appearance is often a critical factor in influencing the user's purchasing decisions. Current state-of-the-art visually-aware recommender systems utilize image features extracted from pre-trained deep convolutional neural networks, however these extremely high-dimensional representations are difficult to interpret, especially in relation to the relatively low number of visual properties that may guide users' decisions. In this paper we propose a novel approach to personalized clothing recommendation that models the dynamics of individual users' visual preferences. By using interpretable image representations generated with a unique feature learning process, our model learns to explain users' prior feedback in terms of their affinity towards specific visual attributes and styles. Our approach achieves state-of-the-art performance on personalized ranking tasks, and the incorporation of interpretable visual features allows for powerful model introspection, which we demonstrate by using an interactive recommendation algorithm and visualizing the rise and fall of fashion trends over time.

Arnau Ramisa, Fei Yan, Francesc Moreno-Noguer et al.

Building upon recent Deep Neural Network architectures, current approaches lying in the intersection of computer vision and natural language processing have achieved unprecedented breakthroughs in tasks like automatic captioning or image retrieval. Most of these learning methods, though, rely on large training sets of images associated with human annotations that specifically describe the visual content. In this paper we propose to go a step further and explore the more complex cases where textual descriptions are loosely related to the images. We focus on the particular domain of News articles in which the textual content often expresses connotative and ambiguous relations that are only suggested but not directly inferred from images. We introduce new deep learning methods that address source detection, popularity prediction, article illustration and geolocation of articles. An adaptive CNN architecture is proposed, that shares most of the structure for all the tasks, and is suitable for multitask and transfer learning. Deep Canonical Correlation Analysis is deployed for article illustration, and a new loss function based on Great Circle Distance is proposed for geolocation. Furthermore, we present BreakingNews, a novel dataset with approximately 100K news articles including images, text and captions, and enriched with heterogeneous meta-data (such as GPS coordinates and popularity metrics). We show this dataset to be appropriate to explore all aforementioned problems, for which we provide a baseline performance using various Deep Learning architectures, and different representations of the textual and visual features. We report very promising results and bring to light several limitations of current state-of-the-art in this kind of domain, which we hope will help spur progress in the field.

Ariadna Quattoni, Arnau Ramisa, Pranava Swaroop Madhyastha et al.

We address the task of annotating images with semantic tuples. Solving this problem requires an algorithm which is able to deal with hundreds of classes for each argument of the tuple. In such contexts, data sparsity becomes a key challenge, as there will be a large number of classes for which only a few examples are available. We propose handling this by incorporating feature representations of both the inputs (images) and outputs (argument classes) into a factorized log-linear model, and exploiting the flexibility of scoring functions based on bilinear forms. Experiments show that integrating feature representations of the outputs in the structured prediction model leads to better overall predictions. We also conclude that the best output representation is specific for each type of argument.