Visual Pivoting for (Unsupervised) Entity Alignment

Authors: Fangyu Liu, Muhao Chen, Dan Roth, Nigel Collier4257-4266

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on benchmark data sets DBP15k and DWY15k show that EVA offers state-of-the-art performance on both monolingual and cross-lingual entity alignment tasks. Furthermore, we discover that images are particularly useful to align long-tail KG entities, which inherently lack the structural contexts that are necessary for capturing the correspondences. Code release: https://github.com/cambridgeltl/eva; project page: http://cogcomp.org/page/publication view/927. In this section, we conduct experiments on two benchmark data sets ( 4.1), under both semiand unsupervised settings ( 4.2). We also provide detailed ablation studies on different model components ( 4.3), and study the impact of incorporating visual representations on long-tail entities ( 4.4).
Researcher Affiliation Academia 1 Language Technology Lab, TAL, University of Cambridge, UK 2 Department of Computer and Information Science, University of Pennsylvania, USA 3 Viterbi School of Engineering, University of Southern California, USA
Pseudocode Yes Algorithm 1: Visual pivot induction.
Open Source Code Yes Code release: https://github.com/cambridgeltl/eva; project page: http://cogcomp.org/page/publication view/927.
Open Datasets Yes The experiments are conducted on DBP15k (Sun, Hu, and Li 2017) and DWY15k (Guo, Sun, and Hu 2019).
Dataset Splits No The paper states 'using 30% of the EA labels for training' and discusses test sets, but does not explicitly provide percentages or counts for a separate validation split or refer to a standard validation split.
Hardware Specification No The paper does not provide specific hardware details such as exact GPU/CPU models, processor types, or memory amounts used for running its experiments.
Software Dependencies No The paper mentions software components like 'Adam W', 'RESNET-152', and 'FASTTEXT' but does not provide specific version numbers for them within the main text.
Experiment Setup Yes The GCN has two layers with input, hidden and output dimensions of 400, 400, 200 respectively. Attribute and relation features are mapped to 100-d. Images are transformed to 2048-d features by RESNET and then mapped to 200-d. For model variants without IL, training is limited to 500 epochs. Otherwise, after the first 500 epochs, IL is conducted for another 500 epochs with the configurations Ke = 5, Ks = 10 as described in 3.2. We train all models using a batch size of 7,500. The models are optimised using Adam W (Loshchilov and Hutter 2019) with a learning rate of 5e-4 and a weight decay of 1e-2.