paper-with-me

홈 › Papers

Ruminating Word Representations with Random Noise Masking

2021-01-01 · Hwiyeol Jo, Byoung-Tak Zhang

We introduce a training method for better word representation and performance, which we call \textbf{GraVeR} (\textbf{Gra}dual \textbf{Ve}ctor \textbf{R}umination). The method is to gradually and iteratively add random noises and bias to word embeddings after training a model, and re-train the model from scratch but initialize with the noised word embeddings. Through the re-training process, some of noises can be compensated and other noises can be utilized to learn better representations. As a result, we can get word representations further fine-tuned and specialized in the task. On six text classification tasks, our method improves model performances with a large gap. When GraVeR is combined with other regularization techniques, it shows further improvements. Lastly, we investigate the usefulness of GraVeR for pretraining by training data.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

text-classificationText ClassificationWord Embeddings

Similar Papers 제목 키워드 기반

Ruminating Word Representations with Random Noised Masker

2019-11-08 · Hwiyeol Jo, Byoung-Tak Zhang

We introduce a training method for both better word representation and performance, which we call GROVER (Gradual Rumination On the Vector with maskERs). The method is to gradually and iteratively add random noises to wo…

text-classificationText ClassificationWord Embeddings

ColorMAE: Exploring data-independent masking strategies in Masked AutoEncoders

2024-07-17 · Carlos Hinojosa, Shuming Liu, Bernard Ghanem

Masked AutoEncoders (MAE) have emerged as a robust self-supervised framework, offering remarkable performance across a wide range of downstream tasks. To increase the difficulty of the pretext task and learn richer visua…

Image ClassificationInstance SegmentationObject DetectionSemantic Segmentation

Do Blind Spots Matter for Word-Referent Mapping? A Computational Study with Infant Egocentric Video

2025-11-13 · Zekai Shi, Zhixi Cai, Kalin Stefanov arxiv

Typically, children start to learn their first words between 6 and 9 months, linking spoken utterances to their visual referents. Without prior knowledge, a word encountered for the first time can be interpreted in count…

Contrastive Learning

Pre-training with Random Orthogonal Projection Image Modeling

2023-10-28 · Maryam Haghighat, Peyman Moghadam, Shaheer Mohamed, Piotr Koniusz

Masked Image Modeling (MIM) is a powerful self-supervised strategy for visual pre-training without the use of labels. MIM applies random crops to input images, processes them with an encoder, and then recovers the masked…

Decoder

Joint Low-level and High-level Textual Representation Learning with Multiple Masking Strategies

2025-05-11 · Zhengmi Tang, Yuto Mitsui, Tomo Miyazaki, Shinichiro Omachi

Most existing text recognition methods are trained on large-scale synthetic datasets due to the scarcity of labeled real-world datasets. Synthetic images, however, cannot faithfully reproduce real-world scenarios, such a…

Contrastive LearningImage Super-ResolutionRepresentation LearningSelf-Supervised Learning+1