Challenge: Existing multimodal pre-training models require large amounts of training data and have huge model sizes, making them impossible to apply in low-resource situations.
Approach: They propose a multi-stage pre-training method which uses information at different granularities from word, phrase to sentence in both texts and images to pre-train a model in stages.
Outcome: The proposed method outperforms the original model in Image-Text Retrieval task and outperformed the original LXMERT model in downstream tasks.

Similar Papers

MSP: Multi-Stage Prompting for Making Pre-trained Language Models Better Translators (2022.acl-long)

Copied to clipboard

Challenge: Prompting has been shown to be a promising approach for applying pre-trained language models to perform downstream tasks.
Approach: They propose a method that divides the translation process into three stages using pre-trained language models.
Outcome: The proposed method significantly improves translation performance of pre-trained language models on three translation tasks.
Complicate Then Simplify: A Novel Way to Explore Pre-trained Models for Text Classification (2022.coling-1)

Copied to clipboard

Challenge: Existing frameworks for text classification employing pre-trained models are constrained by the difficulty of the task.
Approach: They propose a framework which implements a two-stage training strategy to fully exploit the knowledge in pre-trained models.
Outcome: The proposed framework outperforms state-of-the-art classification models on six text classification corpora.
Multi-Stage Multi-Modal Pre-Training for Automatic Speech Recognition (2024.lrec-main)

Copied to clipboard

Challenge: Existing methods for pre-training for automatic speech recognition (ASR) focus on single-stage pre-train followed by fine-tuning on downstream task.
Approach: They propose a multi-modal pre-training method that combines unsupervised pre-training with translation-based supervised mid-training.
Outcome: The proposed method improves WERs by 38.45% over baselines on both Librispeech and SUPERB.
MGDoc: Pre-training with Multi-granular Hierarchy for Document Image Understanding (2022.emnlp-main)

Copied to clipboard

Challenge: Existing methods learn features from word-level or region-level but fail to consider both simultaneously.
Approach: They propose a multi-modal multi-granular pre-training framework that encodes page-level, region-level and word-level information at the same time.
Outcome: The proposed model learns features from word-level and region-level but fails to consider both simultaneously.
Multi-Stage Pre-training for Low-Resource Domain Adaptation (2020.emnlp-main)

Copied to clipboard

Challenge: Existing approaches to transfer learning target data to in-domain text . prior work has adapted pre-trained LMs to specific domains .
Approach: They extend the vocabulary of a pretrained language model with domain-specific terms to create synthetic tasks that help it transfer to downstream tasks.
Outcome: The proposed approaches show significant performance gains on extractive reading comprehension, document ranking and duplicate question detection tasks.
Different Strokes for Different Folks: Investigating Appropriate Further Pre-training Approaches for Diverse Dialogue Tasks (2021.emnlp-main)

Copied to clipboard

Challenge: Pre-trained models can be fine-tuned on domain-specific unlabeled data . however, most further pre-training works just keep running the conventional pre- training task .
Approach: They propose to add a further pre-training phase to the model to improve downstream tasks . they propose to use a domain-adaptive pre-tuning phase to fine-tune the models on unlabeled data .
Outcome: The proposed method improves multiple task-oriented dialogue downstream tasks.
Advances in Pre-Training Distributed Word Representations (L18-1)

Copied to clipboard

Challenge: Pre-trained word representations are a building block of many Natural Language Processing and Machine Learning applications.
Approach: They propose to combine known tricks and a set of publicly available pre-trained word vector representations to train high-quality representations.
Outcome: The proposed models outperform the current state of the art on a number of tasks while maintaining a high training speed to scale to massive amount of data.
An Empirical Investigation Towards Efficient Multi-Domain Language Model Pre-training (2020.emnlp-main)

Copied to clipboard

Challenge: Pre-training large language models is a standard practice in the natural language processing community.
Approach: They propose to use elastic weight consolidation to mitigate catastrophic forgetting when pre-trained large language models are evaluated on generic benchmarks.
Outcome: The proposed model achieves state-of-the-art on out-of domain tasks with minimal pre-training . elastic weight consolidation provides best overall scores yielding only a 0.33% drop in performance across seven generic tasks while remaining competitive in bio-medical tasks.
Cross-lingual Visual Pre-training for Multimodal Machine Translation (2021.eacl-main)

Copied to clipboard

Challenge: Pre-trained language models have been shown to improve performance in many natural language tasks.
Approach: They propose to combine cross-lingual and visual pre-training to learn visually-grounded cross-linguistic representations using masked region classification and three-way parallel vision & language corpora.
Outcome: The proposed models obtain state-of-the-art performance when fine-tuned for multimodal machine translation.
Teaching the Pre-trained Model to Generate Simple Texts for Text Simplification (2023.findings-acl)

Copied to clipboard

Challenge: Existing strategies to teach pre-trained models to generate simple texts are inadequate.
Approach: They propose a continued pre-training strategy to teach pre-trained models to generate simple texts by randomly masking text spans in ordinary texts.
Outcome: The proposed strategy improves on lexical simplification, sentence simplification and document-level simplification tasks over existing models.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations