Microsoft Technology Licensing, LLC

Wenhan Wang

Robin Abraham

Liang Du

Yuxiong He

Minjia Zhang

Fahimeh Raja

Jianfeng Gao

Xiaodong Liu

Mingyang Xu

Yijian Xiang

Sam Daniel Gammon

Dustin James Stewart

Stuart Richard Long

Timothy Earnheart

Henrique Dantas

Manqing Mao

Jedediah Miller Singer

Sacha Arozarena Valladare

Julia Chen

Paishun Ting

Jianzhe Lin

Samyam Rajbhandari

Lipsa Patnaik

The present disclosure relates to methods and systems for using large language models to support research activities. The methods and systems include a copilot engine that creates input prompts to provide to the large language model to use in generating responses to input messages. The copilot engine infers an intent of the input messages and sends the intent with the input message in the input prompt to the large language model. The large language model generates different types of responses for different intents.

Research activities through conversational user experience using multi-modal large pretrained models

The present disclosure relates to methods and systems for sharing with a plurality of users a chat session that uses large language models to provide responses for input messages received for the chat session. The methods and systems provide access to the chat session to the users and update the chat session in response to any changes made to the chat session by any of the users. The methods and systems allow the users to resume the chat session at a future time using the chat session history.

Copilot for multi-user, multi-step collaborations

Systems, methods, and computer-executable instructions for determining a computation schedule for a recurrent neural network (RNN). A matrix multiplication (MM) directed-acyclic graph (DAG) is received for the RNN. Valid phased computation schedules for the RNN are generated. Each of the valid phase computation schedule includes an ordering of MM operations. For each of the plurality of valid phased computation schedules, each of the MM operations is partitioned to processor cores based on L3 cache to L2 cache data movement. The RNN is executed based on the valid phased computation schedules. A final computation schedule is stored. The final computation schedule is used for future executions of the RNN.

Deep learning model scheduling

Systems, methods, and computer-executable instructions for approximating a softmax layer are disclosed. A small world graph that includes a plurality of nodes is constructed for a vocabulary of a natural language model. A context vector is transformed. The small world graph is searched using the transformed context vector to identify a top-K hypothesis. A distance from the context vector for each of the top-K hypothesis is determined. The distance is transformed to an original inner product space. A softmax distribution is computed for the softmax layer over the inner product space of the top-K hypothesis. The softmax layer is useful for determining a next word in a speech recognition or machine translation.

Graph representations for identifying a next word

Growing community of inventors

Redmond, WA, United States of America

Wenhan Wang