IP Library Granted Patent US 12675644
Granted Patent B2
US 12675644 · App. 17/807,744 · Granted Jul 7, 2026

Task-specific language sets for multilingual learning

Inventors: Tejas Indulal Dhamecha (Bangalore, IN); Samarth Bharadwaj (Bangalore, IN); Rudra Murthy Venkataramana (Bangalore, IN); Jaydeep Sen (Bangalore, IN); Anupama Ray (Bangalore, IN)
Assignee: International Business Machines Corporation
G06F40/40
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12675644
App. No.
17/807,744
Granted
Jul 7, 2026
Kind
B2
Abstract

A method, a structure, and a computer system for multilingual learning. The exemplary embodiments may include training, for each language in a set of two or more languages, a model for a task and identifying one or more important words appearing in at least two of the models. The exemplary embodiments may further include weighting one or more conflicts and one or more overlaps between the one or more important words, as well as generating a cluster of at least two languages of the set based on an aggregate of the weighting.

Claims (34)

1 . A method for multilingual learning, the method comprising:

training, for each language in a set of two or more languages, a model for a task;

identifying one or more important words appearing in at least two of the models;

weighting one or more negative conflicts and one or more positive overlaps between the one or more important words, wherein the one or more negative conflicts connote the one or more important words having a negative interference with joint learning;

generating a cluster of at least two languages of the set based on an aggregate of the weighting; and

utilizing the cluster in one or more higher-order tasks selected from the group consisting of text classification, natural language inference, paraphrase detection, and Multilingual LM.

2 . The method of claim 1 , wherein the two or more languages are labelled for the task.

3 . The method of claim 1 , wherein the training is via fine-tuning a pretrained model.

4 . The method of claim 1 , wherein the task is selected from the group consisting of sentiment classification, natural language interference, and part-of-speech tagging.

5 . The method of claim 1 , wherein the weighting the one or more conflicts and the one or more overlaps is based on the task.

6 . The method of claim 1 , wherein the identifying one or more important words appearing in at least two of the models is based on the task.

7 . A computer program product for multilingual learning, the computer program product comprising:

one or more non-transitory computer-readable storage media and program instructions stored on the one or more non-transitory computer-readable storage media capable of performing a method, the method comprising:

training, for each language in a set of two or more languages, a model for a task;

identifying one or more important words appearing in at least two of the models;

weighting one or more negative conflicts and one or more positive overlaps between the one or more important words, wherein the one or more negative conflicts connote the one or more important words having a negative interference with joint learning;

generating a cluster of at least two languages of the set based on an aggregate of the weighting; and

utilizing the cluster in one or more higher-order tasks selected from the group consisting of text classification, natural language inference, paraphrase detection, and Multilingual LM.

8 . The computer program product of claim 7 , wherein the two or more languages are labelled for the task.

9 . The computer program product of claim 7 , wherein the training is via fine-tuning a pretrained model.

10 . The computer program product of claim 7 , wherein the task is selected from the group consisting of sentiment classification, natural language interference, and part-of-speech tagging.

11 . The computer program product of claim 7 , wherein the weighting the one or more conflicts and the one or more overlaps is based on the task.

12 . The computer program product of claim 7 , wherein the identifying one or more important words appearing in at least two of the models is based on the task.

13 . A computer system for multilingual learning, the system comprising:

one or more computer processors, one or more computer-readable storage media, and program instructions stored on the one or more of the computer-readable storage media for execution by at least one of the one or more processors capable of performing a method, the method comprising:

training, for each language in a set of two or more languages, a model for a task;

identifying one or more important words appearing in at least two of the models;

weighting one or more negative conflicts and one or more positive overlaps between the one or more important words, wherein the one or more negative conflicts connote the one or more important words having a negative interference with joint learning;

generating a cluster of at least two languages of the set based on an aggregate of the weighting; and

utilizing the cluster in one or more higher-order tasks selected from the group consisting of text classification, natural language inference, paraphrase detection, and Multilingual LM.

14 . The computer system of claim 13 , wherein the two or more languages are labelled for the task.

15 . The computer system of claim 13 , wherein the training is via fine-tuning a pretrained model.

16 . The computer system of claim 13 , wherein the task is selected from the group consisting of sentiment classification, natural language interference, and part-of-speech tagging.

17 . The computer system of claim 13 , wherein the identifying one or more important words appearing in at least two of the models and the weighting the one or more conflicts and the one or more overlaps is based on the task.