e-CLIP: Large-Scale Vision-Language Representation Learning in E-commerce

Understanding vision and language representations of product content is vital for search and recommendation applications in e-commerce. As a backbone for online shopping platforms and inspired by the recent success in representation learning research, we propose a contrastive learning framework that aligns language and visual models using unlabeled raw product text and images. We present techniques we used to train large-scale representation learning models and share solutions that address domain-specific challenges. We study the performance using our pre-trained model as backbones for diverse downstream tasks, including category classification, attribute extraction, product matching, product clustering, and adult product recognition. Experimental results show that our proposed method outperforms the baseline in each downstream task regarding both single modality and multiple modalities.

Learning a Word-Level Language Model with Sentence-Level Noise Contrastive Estimation for Contextual Sentence Probability Estimation

Inferring the probability distribution of sentences or word sequences is a key process in natural language processing. While word-level language models (LMs) have been widely adopted for computing the joint probabilities of word sequences, they have difficulty in capturing a context long enough for sentence probability estimation (SPE). To overcome this, recent studies introduced training methods using sentence-level noise-contrastive estimation (NCE) with recurrent neural networks (RNNs). In this work, we attempt to extend it for contextual SPE, which aims to estimate a conditional sentence probability given a previous text. The proposed NCE samples negative sentences independently of a previous text so that the trained model gives higher probabilities to the sentences that are more consistent with \textcolor{blue}{the} context. We apply our method to a simple word-level RNN LM to focus on the effect of the sentence-level NCE training rather than on the network architecture. The quality of estimation was evaluated against multiple-choice cloze-style questions including both human and automatically generated questions. The experimental results show that the proposed method improved the SPE quality for the word-level RNN LM.

Image-to-Image Retrieval by Learning Similarity between Scene Graphs

As a scene graph compactly summarizes the high-level content of an image in a structured and symbolic manner, the similarity between scene graphs of two images reflects the relevance of their contents. Based on this idea, we propose a novel approach for image-to-image retrieval using scene graph similarity measured by graph neural networks. In our approach, graph neural networks are trained to predict the proxy image relevance measure, computed from human-annotated captions using a pre-trained sentence similarity model. We collect and publish the dataset for image relevance measured by human annotators to evaluate retrieval algorithms. The collected dataset shows that our method agrees well with the human perception of image similarity than other competitive baselines.

A Bi-directional Transformer for Musical Chord Recognition

Chord recognition is an important task since chords are highly abstract and descriptive features of music. For effective chord recognition, it is essential to utilize relevant context in audio sequence. While various machine learning models such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) have been employed for the task, most of them have limitations in capturing long-term dependency or require training of an additional model. In this work, we utilize a self-attention mechanism for chord recognition to focus on certain regions of chords. Training of the proposed bi-directional Transformer for chord recognition (BTC) consists of a single phase while showing competitive performance. Through an attention map analysis, we have visualized how attention was performed. It turns out that the model was able to divide segments of chords by utilizing adaptive receptive field of the attention mechanism. Furthermore, it was observed that the model was able to effectively capture long-term dependencies, making use of essential information regardless of distance.