|
Work item:
|
F.TE-CG
|
|
Subject/title:
|
Technical requirements and evaluation methods of AI based code generation in multimedia applications
|
|
Status:
|
[Carried to next study period]
|
|
Approval process:
|
AAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
-
|
|
Liaison:
|
-
|
|
Supporting members:
|
-
|
|
Summary:
|
This recommendation aims to establish comprehensive specifications for the production of code generation based on AI within the multimedia domain. It covers the technical requirements and evaluation criteria for code understanding, code generation and completion, code conversion, unit test case generation, model maturity, service maturity and so on. In doing so, it seeks to address the significant challenges that impede the implementation of this technology, including inconsistent definitions of capabilities, absence of standardized evaluation criteria, incomplete evaluation indicator systems, inequities in ranking assessments, and a lack of unified requirements for segmented scene capabilities.
This recommendation aims to provide guidance to producer enterprises in their product research endeavors, standardize the construction processes of code generation models or tools, and furnish user enterprises with valuable selection references for products in this domain. Ultimately, this initiative intends to accelerate the industrial implementation of code generation based on AI in the multimedia sector, enhancing the efficiency and user experience of multimedia applications, and heralding a new era of innovation and advancement in the multimedia software development landscape.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2024-06-20 17:53:03
|
|
Last update:
2024-09-20 10:46:22
|
|
