paper-with-me

홈 › Papers

Cultural and Geographical Influences on Image Translatability of Words across Languages

2021-06-01 · NAACL 2021 4 · Nikzad Khani, Isidora Tourni, Mohammad Sadegh Rasooli, Chris Callison-Burch, Derry Tanti Wijaya

Neural Machine Translation (NMT) models have been observed to produce poor translations when there are few/no parallel sentences to train the models. In the absence of parallel data, several approaches have turned to the use of images to learn translations. Since images of words, e.g., horse may be unchanged across languages, translations can be identified via images associated with words in different languages that have a high degree of visual similarity. However, translating via images has been shown to improve upon text-only models only marginally. To better understand when images are useful for translation, we study image translatability of words, which we define as the translatability of words via images, by measuring intra- and inter-cluster similarities of image representations of words that are translations of each other. We find that images of words are not always invariant across languages, and that language pairs with shared culture, meaning having either a common language family, ethnicity or religion, have improved image translatability (i.e., have more similar images for similar words) compared to its converse, regardless of their geographic proximity. In addition, in line with previous works that show images help more in translating concrete words, we found that concrete words have improved image translatability compared to abstract ones.

📄 PDF Abstract BibTeX

Code (1)

nikzadkhani/MMID-CNN-Analysis 공식 구현

Tasks

Cultural Vocal Bursts Intensity PredictionLow Resource Neural Machine TranslationLow-Resource Neural Machine TranslationMachine TranslationMultilingual NLPMultimodal Machine TranslationNMTTranslation

Methods 이 논문이 사용한 방법론

Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Transformer A Transformer is a model architecture that eschews recurrence and instead relies entirely on an [attention…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음
Weight Decay 설명 없음

Similar Papers 제목 키워드 기반

World Wide Models: Literary Tools for Cultural AI

2026-07-02 · Nina Begus arxiv

LLMs stage a new form of cultural encounter that is massive, automated, and monolingual. Literary disciplines have always negotiated cultural struggles with comparative reading of literature, narratological and poetic an…

Measuring Cultural Relativity of Emotional Valence and Arousal using Semantic Clustering and Twitter

2013-04-28 · Eugene Yuta Bann, Joanna J. Bryson

Researchers since at least Darwin have debated whether and to what extent emotions are universal or culture-dependent. However, previous studies have primarily focused on facial expressions and on a limited set of emotio…

ClusteringCultural Vocal Bursts Intensity Prediction

Networks and Identity Drive Geographic Properties of the Diffusion of Linguistic Innovation

2022-02-10 · Aparna Ananthasubramaniam, David Jurgens, Daniel M. Romero

Adoption of cultural innovation (e.g., music, beliefs, language) is often geographically correlated, with adopters largely residing within the boundaries of relatively few well-studied, socially significant areas. These …

Cultural Vocal Bursts Intensity Prediction

DisenPOI: Disentangling Sequential and Geographical Influence for Point-of-Interest Recommendation

2022-10-29 · Yifang Qin, Yifan Wang, Fang Sun, Wei Ju 외

Point-of-Interest (POI) recommendation plays a vital role in various location-aware services. It has been observed that POI recommendation is driven by both sequential and geographical influences. However, since there is…

Contrastive LearningDisentanglement

BAR-Analytics: A Web-based Platform for Analyzing Information Spreading Barriers in News: Comparative Analysis Across Multiple Barriers and Events

2025-03-31 · Abdul Sittar, Dunja Mladenic, Alenka Gucek, Marko Grobelnik

This paper presents BAR-Analytics, a web-based, open-source platform designed to analyze news dissemination across geographical, economic, political, and cultural boundaries. Using the Russian-Ukrainian and Israeli-Pales…

ArticlesSentiment Analysis