2021
DOI: 10.1609/aaai.v35i9.16974
|View full text |Cite
|
Sign up to set email alerts
|

LightXML: Transformer with Dynamic Negative Sampling for High-Performance Extreme Multi-label Text Classification

Abstract: Extreme multi-label text classification(XMC) is a task for finding the most relevant labels from a large label set. Nowadays deep learning-based methods have shown significant success in XMC. However, the existing methods (e.g., AttentionXML and X-Transformer etc) still suffer from 1) combining several models to train and predict for one dataset, and 2) sampling negative labels statically during the process of training label ranking model, which will harm the performance and accuracy of model. To address the a… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
31
0

Year Published

2022
2022
2023
2023

Publication Types

Select...
3
2
1

Relationship

0
6

Authors

Journals

citations
Cited by 69 publications
(51 citation statements)
references
References 8 publications
0
31
0
Order By: Relevance
“…For Wikipedia-500K, Amazon-670K, and Amazon-3M, we use the same experimental setup (i.e. raw input text, sparse features and train-test split) as existing deep XMC methods [31,33,18,7]. For LF-AmazonTitles-131K, we use the experimental setup provided in the extreme classification repository [5].…”
Section: Resultsmentioning
confidence: 99%
See 4 more Smart Citations
“…For Wikipedia-500K, Amazon-670K, and Amazon-3M, we use the same experimental setup (i.e. raw input text, sparse features and train-test split) as existing deep XMC methods [31,33,18,7]. For LF-AmazonTitles-131K, we use the experimental setup provided in the extreme classification repository [5].…”
Section: Resultsmentioning
confidence: 99%
“…BERT-OvA-1 Method P@1 P@3 P@5 R@10 R@20 R@100 P@1 P@3 P@5 R@10 R@20 R@100 Amazon-670K LF-AmazonTitles-131K Comparison on XMC benchmarks Table 1 compares our method with leading XMC methods such as DiSMEC [2], Parabel [25], XR-Linear [32], Bonsai [19], Slice [16], Astec [9], GlaS [13], AttentionXML [31], LightXML [18], XR-Transformer [33], and Overlap-XMC [22]. Most baseline results are obtained from their respective papers when available and otherwise taken from results reported in [31,33] and extreme classification repository [5].…”
Section: Methodsmentioning
confidence: 99%
See 3 more Smart Citations