Invention Application
- Patent Title: SYSTEMS AND METHODS FOR SEMANTIC CODE SEARCH
-
Application No.: US17531591Application Date: 2021-11-19
-
Publication No.: US20220374595A1Publication Date: 2022-11-24
- Inventor: Akhilesh Deepak Gotmare , Junnan Li , Shafiq Rayhan Joty , Chu Hong Hoi
- Applicant: salesforce.com, inc.
- Applicant Address: US CA San Francisco
- Assignee: salesforce.com, inc.
- Current Assignee: salesforce.com, inc.
- Current Assignee Address: US CA San Francisco
- Main IPC: G06F40/226
- IPC: G06F40/226 ; G06F40/40 ; G06F40/30 ; G06F40/151

Abstract:
Embodiments described herein provides a contrastive learning framework that leverages hard negative examples, that are mined globally from the entire training corpus for a given query to improve the quality of code and natural language representations. Specifically, similar examples from the training corpus are extracted and used as hard negatives in an online manner during training while keeping the minibatch construction random.
Information query