Sign in

Continuity of care is crucial to ensuring positive health outcomes for patients discharged from an inpatient hospital setting, and improved information sharing via discharge summaries can help. When patients are discharged, they often require further actions to be taken by their primary care provider (PCP), who manages their long-term health…

94% accurate automated response model may have seemed like a good result, but after careful testing, we realized that it’s not. When it comes to a standard use case we actually can not afford to mistake. …

Motivated by the results of our latest experiment, we decided to make another similar one. The difference is that in half of the cases of the training set, some user information as customer_name, phone, zip_code will be given before the dialogue begins. …

Most businesses focus on a relatively small number of products and services, and their customer queries are limited to a few topics/intents.
For example, the Airline customer service queries will mostly be limited to booking a ticket, canceling a reservation, and checking the flight status. Despite the small number of intents…

Today, most Machine Learning driven chatbots rely on Intent Classification models to extract the answer from the database for a given question. The model tends to work well during testing but may lack in a real conversation. The reason for that is “Context”. The thing is that in a real…

Asapp has recently released Action based Conversational Dataset for Action State Tracking (AST) and Cascading Dialogue Success (CDS) research.

ABCD (https://github.com/asappresearch/abcd) dataset consists of 10,042 conversations between agents and customers. It also contains sets of actions that agents perform during the conversation in order to process some information or collect…

AirDialogue is one of the largest domain-based conversational datasets provided by google — https://github.com/google/airdialogue

In this work, we have trained the model that is able to go through long conversations between airline agents and customers, then extract the required information from these dialogs. The goal was to achieve very high…

Dialogue state tracking is one of the biggest problems in Conversational Technologies Research. Recent progress in dialogue generation has inspired several studies on dialogue systems that are capable to accomplish tasks through natural language interactions. A promising direction among these studies is the use of pre-trained models like BERT for…

Transfer learning, where a model is first pre-trained on a data-rich task before being fine-tuned on a downstream task, has emerged as a powerful technique in natural language processing (NLP). The effectiveness of transfer learning has given rise to a diversity of approaches, methodology, and practice. In this work, we…

AILabs

Conversational AI Research company. https://ailabs.us

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store