Salesforce.com

About: Salesforce.com is a based out in . It is known for research contribution in the topics: User interface & Object (computer science). The organization has 2418 authors who have published 2775 publications receiving 63956 citations.

Custom entities and fields in a multi-tenant database system

Abstract: Systems and methods for hosting variable schema data such as dynamic tables and columns in a fixed physical database schema. Standard objects, such as tables are provided for use by multiple tenants or organizations in a multi-tenant database system. Each organization may add or define custom fields for inclusion in a standard object. Custom fields for multiple tenants are stored in a single field within the object data structure, and this single field may contain different data types for each tenant. Indexing columns are also provided, wherein a tenant may designate a field for indexing. Data values for designated fields are copied to an index column, and each index column may include multiple data types. Each organization may also define custom objects including custom fields and indexing columns. Custom objects for multiple tenants are stored in a single custom object data structure. The primary key values for the single custom object table are globally unique, but also include an object-specific identifier which may be re-used among different entities.

Position-aware Attention and Supervised Data Improve Slot Filling

Abstract: Organized relational knowledge in the form of “knowledge graphs” is important for many applications. However, the ability to populate knowledge bases with facts automatically extracted from documents has improved frustratingly slowly. This paper simultaneously addresses two issues that have held back prior work. We first propose an effective new model, which combines an LSTM sequence model with a form of entity position-aware attention that is better suited to relation extraction. Then we build TACRED, a large (119,474 examples) supervised relation extraction dataset obtained via crowdsourcing and targeted towards TAC KBP relations. The combination of better supervised data and a more appropriate high-capacity model enables much better relation extraction performance. When the model trained on this new dataset replaces the previous relation extraction component of the best TAC KBP 2015 slot filling system, its F1 score increases markedly from 22.2% to 26.7%.

Learned in translation: contextualized word vectors

Abstract: Computer vision has benefited from initializing multiple deep layers with weights pretrained on large supervised training sets like ImageNet. Natural language processing (NLP) typically sees initialization of only the lowest layer of deep models with pretrained word vectors. In this paper, we use a deep LSTM encoder from an attentional sequence-to-sequence model trained for machine translation (MT) to contextualize word vectors. We show that adding these context vectors (CoVe) improves performance over using only unsupervised word and character vectors on a wide variety of common NLP tasks: sentiment analysis (SST, IMDb), question classification (TREC), entailment (SNLI), and question answering (SQuAD). For fine-grained sentiment analysis and entailment, CoVe improves performance of our baseline models to the state of the art.

Dynamic memory networks for visual and textual question answering

Abstract: Neural network architectures with memory and attention mechanisms exhibit certain reasoning capabilities required for question answering. One such architecture, the dynamic memory network (DMN), obtained high accuracy on a variety of language tasks. However, it was not shown whether the architecture achieves strong results for question answering when supporting facts are not marked during training or whether it could be applied to other modalities such as images. Based on an analysis of the DMN, we propose several improvements to its memory and input modules. Together with these changes we introduce a novel input module for images in order to be able to answer visual questions. Our new DMN+ model improves the state of the art on both the Visual Question Answering dataset and the bAbI-10k text question-answering dataset without supporting fact supervision.

Learned in Translation: Contextualized Word Vectors

Abstract: Computer vision has benefited from initializing multiple deep layers with weights pretrained on large supervised training sets like ImageNet. Natural language processing (NLP) typically sees initialization of only the lowest layer of deep models with pretrained word vectors. In this paper, we use a deep LSTM encoder from an attentional sequence-to-sequence model trained for machine translation (MT) to contextualize word vectors. We show that adding these context vectors (CoVe) improves performance over using only unsupervised word and character vectors on a wide variety of common NLP tasks: sentiment analysis (SST, IMDb), question classification (TREC), entailment (SNLI), and question answering (SQuAD). For fine-grained sentiment analysis and entailment, CoVe improves performance of our baseline models to the state of the art.