paper-with-me

홈 › Papers

Using Two Losses and Two Datasets Simultaneously to Improve TempoWiC Accuracy

2022-12-15 · Mohammad Javad Pirhadi, Motahhare Mirzaei, Sauleh Eetemadi

WSD (Word Sense Disambiguation) is the task of identifying which sense of a word is meant in a sentence or other segment of text. Researchers have worked on this task (e.g. Pustejovsky, 2002) for years but it's still a challenging one even for SOTA (state-of-the-art) LMs (language models). The new dataset, TempoWiC introduced by Loureiro et al. (2022b) focuses on the fact that words change over time. Their best baseline achieves 70.33% macro-F1. In this work, we use two different losses simultaneously to train RoBERTa-based classification models. We also improve our model by using another similar dataset to generalize better. Our best configuration beats their best baseline by 4.23% and reaches 74.56% macroF1.

📄 PDF Abstract BibTeX arXiv:2212.07669

Code (0)

등록된 구현이 없습니다.

Tasks

SentenceVocal Bursts Valence PredictionWord Sense Disambiguation

Similar Papers 제목 키워드 기반

TempoWiC: An Evaluation Benchmark for Detecting Meaning Shift in Social Media

2022-09-15 · COLING 2022 10 · Daniel Loureiro, Aminette D'Souza, Areej Nasser Muhajab, Isabella A. White 외

Language evolves over time, and word meaning changes accordingly. This is especially true in social media, since its dynamic nature leads to faster semantic shifts, making it challenging for NLP models to deal with new c…

Using Deep Mixture-of-Experts to Detect Word Meaning Shift for TempoWiC

2022-11-07 · Ze Chen, Kangxu Wang, Zijian Cai, Jiewen Zheng 외

This paper mainly describes the dma submission to the TempoWiC task, which achieves a macro-F1 score of 77.05% and attains the first place in this task. We first explore the impact of different pre-trained language model…

Data AugmentationMixture-of-ExpertsPOS

Revisiting Contrastive Learning through the Lens of Neighborhood Component Analysis: an Integrated Framework

2021-12-08 · Ching-Yun Ko, Jeet Mohapatra, Sijia Liu, Pin-Yu Chen 외

As a seminal tool in self-supervised representation learning, contrastive learning has gained unprecedented attention in recent years. In essence, contrastive learning aims to leverage pairs of positive and negative samp…

Contrastive LearningRepresentation Learning

BD-KD: Balancing the Divergences for Online Knowledge Distillation

2022-12-25 · Ibtihel Amara, Nazanin Sepahvand, Brett H. Meyer, Warren J. Gross 외

We address the challenge of producing trustworthy and accurate compact models for edge devices. While Knowledge Distillation (KD) has improved model compression in terms of achieving high accuracy performance, calibratio…

Knowledge DistillationModel CompressionTransfer Learning

Principled Approaches for Learning to Defer with Multiple Experts

2023-10-23 · Anqi Mao, Mehryar Mohri, Yutao Zhong

We present a study of surrogate losses and algorithms for the general problem of learning to defer with multiple experts. We first introduce a new family of surrogate losses specifically tailored for the multiple-expert …