paper-with-me

홈 › Papers

MIPE: A Metric Independent Pipeline for Effective Code-Mixed NLG Evaluation

2021-07-24 · EMNLP (Eval4NLP) 2021 11 · Ayush Garg, Sammed S Kagi, Vivek Srivastava, Mayank Singh

Code-mixing is a phenomenon of mixing words and phrases from two or more languages in a single utterance of speech and text. Due to the high linguistic diversity, code-mixing presents several challenges in evaluating standard natural language generation (NLG) tasks. Various widely popular metrics perform poorly with the code-mixed NLG tasks. To address this challenge, we present a metric independent evaluation pipeline MIPE that significantly improves the correlation between evaluation metrics and human judgments on the generated code-mixed text. As a use case, we demonstrate the performance of MIPE on the machine-generated Hinglish (code-mixing of Hindi and English languages) sentences from the HinGE corpus. We can extend the proposed evaluation strategy to other code-mixed language pairs, NLG tasks, and evaluation metrics with minimal to no effort.

📄 PDF Abstract BibTeX arXiv:2107.11534

Code (0)

등록된 구현이 없습니다.

Tasks

Diversitynlg evaluationText Generation

Similar Papers 제목 키워드 기반

Improving Paratope and Epitope Prediction by Multi-Modal Contrastive Learning and Interaction Informativeness Estimation

2024-05-31 · Zhiwei Wang, Yongkang Wang, Wen Zhang

Accurately predicting antibody-antigen binding residues, i.e., paratopes and epitopes, is crucial in antibody design. However, existing methods solely focus on uni-modal data (either sequence or structure), disregarding …

Antibody-antigen binding predictionContrastive LearningInformativeness

Multiple Invertible and Partial-Equivariant Function for Latent Vector Transformation to Enhance Disentanglement in VAEs

2025-02-06 · Hee-Jun Jung, Jaehyoung Jeong, Kangil Kim

Disentanglement learning is a core issue for understanding and re-using trained information in Variational AutoEncoder (VAE), and effective inductive bias has been reported as a key factor. However, the actual implementa…

DisentanglementInductive Bias

Learning Descriptive Image Captioning via Semipermeable Maximum Likelihood Estimation

2023-09-21 · NeurIPS 2023 11

Image captioning aims to describe visual content in natural language. As 'a picture is worth a thousand words', there could be various correct descriptions for an image. However, with maximum likelihood estimation as the…

Distantly Supervised Named Entity Recognition using Positive-Unlabeled Learning

2019-06-04 · ACL 2019 7 · Minlong Peng, Xiaoyu Xing, Qi Zhang, Jinlan Fu 외

In this work, we explore the way to perform named entity recognition (NER) using only unlabeled data and named entity dictionaries. To this end, we formulate the task as a positive-unlabeled (PU) learning problem and acc…

named-entity-recognitionNamed Entity RecognitionNamed Entity Recognition (NER)NER+1

Detection of Metallo-\b{eta}-Lactamases-Encoding Genes Among Clinical Isolates of Escherichia Coli in a Tertiary Care Hospital, Malaysia

2019-04-10 · Fazlul Mkk, Deepthi S, Farzana Y, Najnin A 외

The multidrug resistant Escherichia coli strains causes multiple clinical infections and has become a rising problem globally. The metallo-\b{eta}-lactamases encoding genes are very severe in gram-negative bacteria espec…

Management