公开(公告)号：US20240428044A1

公开(公告)日：2024-12-26

申请号：US18497395

申请日：2023-10-30

Applicant: Salesforce, Inc.

Inventor： Ye Liu ,  Semih Yavuz ,  Meghana Moorthy Bhat ,  Rui Meng ,  Shafiq Joty ,  Caiming Xiong ,  Yingbo Zhou

IPC: G06N3/006 ,  G06N3/0455

Abstract: Embodiments described herein provide a framework that integrates a retriever model and the LLM to feed retrieved passages to an LLM to generate an answer conditioned on the retrieved passages in response to a query. For example, in one embodiment, a single-round approach is implemented, which involves directly transmitting the retrieved passages to the LLM. For another example, a multi-round methodology is implemented, which involves initially presenting the retrieved passages to the Language Model, collecting its responses, and then adjusting our interaction with the Language Model based on this acquired feedback.

公开(公告)号：US12105744B2

公开(公告)日：2024-10-01

申请号：US18059691

申请日：2022-11-29

Applicant: Salesforce, Inc.

Inventor： Ye Liu ,  Semih Yavuz ,  Yingbo Zhou ,  Rui Meng

IPC: G06F16/00 ,  G06F16/33 ,  G06F16/332 ,  G06F40/205 ,  G06F40/295 ,  G06F40/30 ,  G06F40/40

CPC classification number: G06F16/3329 ,  G06F16/3344 ,  G06F40/205 ,  G06F40/295 ,  G06F40/30 ,  G06F40/40

Abstract: Embodiments described herein provide a semantic parsing framework which may be referred to as Uni-Parser. The Uni-Parser framework may be applied to question answering on both knowledge bases and databases. The three main stages of the Uni-Parser framework are enumeration, ranking, and generation. At the enumeration stage, primitives are enumerated based on matching the question to the data structure. After enumerating primitives, the Uni-Parser framework may rank the primitives used a trained ranker model. The top ranked primitives may then be used as inputs to a generator which is a learned sequence to sequence model which produces a logical form.

公开(公告)号：US20240202530A1

公开(公告)日：2024-06-20

申请号：US18303313

申请日：2023-04-19

Applicant: Salesforce Inc.

Inventor： Rui Meng ,  Yingbo Zhou ,  Ye Liu ,  Semih Yavuz ,  Ning Yu

IPC: G06N3/084 ,  G06F40/20 ,  G06F40/40 ,  G06N3/0455 ,  G06N3/088

CPC classification number: G06N3/084 ,  G06F40/20 ,  G06F40/40 ,  G06N3/0455 ,  G06N3/088

Abstract: Embodiments described herein provide systems and methods for training a text retrieval model. A system may generate queries associated with provided documents. The queries may be generated in one or more different manners. Examples of query generation may include extracting relevant spans of text from the documents, prompting a language model for a topic, title, abstractive summary, and/or extractive summary based on the documents. Metadata such as title or other HTML tags may be used as queries. Using the one or more queries, the text retrieval model may be trained using contrastive learning, using the generated query, and positive and negative sample documents. A fine-tuning training phase may be performed using domain-specific data which may also be done with generated query pairs, or may be done in a supervised fashion with provided queries. The text retrieval model may be used to locate documents given an input query.