公开(公告)号：US20240086436A1

公开(公告)日：2024-03-14

申请号：US18513981

申请日：2023-11-20

Applicant: Google LLC

Inventor： Thomas Müller ,  Jonathan Herzig ,  Pawel Nowak ,  Julian Eisenschlos ,  Francesco Piccinno ,  Syrine Krichene

IPC: G06F16/332 ,  G06F40/20 ,  G06F40/284 ,  G06F40/35 ,  G06N3/08

CPC classification number: G06F16/3329 ,  G06F40/20 ,  G06F40/284 ,  G06F40/35 ,  G06N3/08

Abstract: Systems and methods for pre-training and fine-tuning of neural-network-based language models to reason directly over tables without generating logical forms. In some examples, a language model can be pre-trained using masked-language modeling tasks synthetically generated from tables pulled from a knowledge corpus. In some examples, the language model may be further pre-trained using pairs of counterfactual statements generated from those tables, and/or one or more statements that compare selected data from those tables. The language model may then be fine-tuned using examples that include only a question, an answer, and a table, allowing fine-tuning examples to be harvested directly from existing benchmark datasets or synthetically generated.

公开(公告)号：US20220309087A1

公开(公告)日：2022-09-29

申请号：US17215465

申请日：2021-03-29

Applicant: Google LLC

Inventor： Thomas Müller ,  Jonathan Herzig ,  Pawel Nowak ,  Julian Eisenschlos ,  Francesco Piccinno ,  Syrine Krichene

IPC: G06F16/332 ,  G06F40/20 ,  G06N3/08

Abstract: Systems and methods for pre-training and fine-tuning of neural-network-based language models to reason directly over tables without generating logical forms. In some examples, a language model can be pre-trained using masked-language modeling tasks synthetically generated from tables pulled from a knowledge corpus. In some examples, the language model may be further pre-trained using pairs of counterfactual statements generated from those tables, and/or one or more statements that compare selected data from those tables. The language model may then be fine-tuned using examples that include only a question, an answer, and a table, allowing fine-tuning examples to be harvested directly from existing benchmark datasets or synthetically generated.

公开(公告)号：US20240386247A1

公开(公告)日：2024-11-21

申请号：US18668086

申请日：2024-05-17

Applicant: Google LLC

Inventor： Zorik Gekhman ,  Jonathan Herzig ,  Roee Aharoni ,  Idan Szpektor

IPC: G06N3/0455

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating, using a language model, a data set for use in performing consistency evaluation for document summaries. For example, the data set can be used to train or evaluate a consistency evaluation neural network.

公开(公告)号：US11868381B2

公开(公告)日：2024-01-09

申请号：US17215465

申请日：2021-03-29

Applicant: Google LLC

Inventor： Thomas Müller ,  Jonathan Herzig ,  Pawel Nowak ,  Julian Eisenschlos ,  Francesco Piccinno ,  Syrine Krichene

IPC: G06F16/332 ,  G06N3/08 ,  G06F40/20 ,  G06F40/284 ,  G06F40/35

CPC classification number: G06F16/3329 ,  G06F40/20 ,  G06F40/284 ,  G06F40/35 ,  G06N3/08

Abstract: Systems and methods for pre-training and fine-tuning of neural-network-based language models to reason directly over tables without generating logical forms. In some examples, a language model can be pre-trained using masked-language modeling tasks synthetically generated from tables pulled from a knowledge corpus. In some examples, the language model may be further pre-trained using pairs of counterfactual statements generated from those tables, and/or one or more statements that compare selected data from those tables. The language model may then be fine-tuned using examples that include only a question, an answer, and a table, allowing fine-tuning examples to be harvested directly from existing benchmark datasets or synthetically generated.