Page: 1 2 3 4 5 6 7 8 9... 52
81 |
Meta Distant Transfer Learning for Pre-trained Language Models ...
|
|
|
|
BASE
|
|
Show details
|
|
83 |
Improving Span Representation for Domain-adapted Coreference Resolution ...
|
|
|
|
BASE
|
|
Show details
|
|
84 |
Temporal Adaptation of BERT and Performance on Downstream Document Classification: Insights from Social Media ...
|
|
|
|
BASE
|
|
Show details
|
|
85 |
An Empirical Study on Multiple Information Sources for Zero-Shot Fine-Grained Entity Typing ...
|
|
|
|
BASE
|
|
Show details
|
|
86 |
Looking for Confirmations: An Effective and Human-Like Visual Dialogue Strategy ...
|
|
|
|
BASE
|
|
Show details
|
|
89 |
CrossVQA: Scalably Generating Benchmarks for Systematically Testing VQA Generalization ...
|
|
|
|
BASE
|
|
Show details
|
|
90 |
Latent Hatred: A Benchmark for Understanding Implicit Hate Speech ...
|
|
|
|
BASE
|
|
Show details
|
|
91 |
HypMix: Hyperbolic Interpolative Data Augmentation ...
|
|
|
|
Abstract:
Anthology paper link: https://aclanthology.org/2021.emnlp-main.776/ Abstract: Interpolation-based regularisation methods for data augmentation have proven to be effective for various tasks and modalities. These methods involve performing mathematical operations over the raw input samples or their latent states representations - vectors that often possess complex hierarchical geometries. However, these operations are performed in the Euclidean space, simplifying these representations, which may lead to distorted and noisy interpolations. We propose HypMix, a novel model-, data-, and modality-agnostic interpolative data augmentation technique operating in the hyperbolic space, which captures the complex geometry of input and hidden state hierarchies better than its contemporaries. We evaluate HypMix on benchmark and low resource datasets across speech, text, and vision modalities, showing that HypMix consistently outperforms state-of-the-art data augmentation techniques. In addition, we demonstrate the use of ...
|
|
Keyword:
Computational Linguistics; Machine Learning; Machine Learning and Data Mining; Natural Language Processing
|
|
URL: https://dx.doi.org/10.48448/z7yd-ks04 https://underline.io/lecture/38086-hypmix-hyperbolic-interpolative-data-augmentation
|
|
BASE
|
|
Hide details
|
|
92 |
STaCK: Sentence Ordering with Temporal Commonsense Knowledge ...
|
|
|
|
BASE
|
|
Show details
|
|
93 |
ExplaGraphs: An Explanation Graph Generation Task for Structured Commonsense Reasoning ...
|
|
|
|
BASE
|
|
Show details
|
|
94 |
Weakly supervised discourse segmentation for multiparty oral conversations ...
|
|
|
|
BASE
|
|
Show details
|
|
95 |
Searching for an Effective Defender: Benchmarking Defense against Adversarial Word Substitution ...
|
|
|
|
BASE
|
|
Show details
|
|
96 |
Progressively Guide to Attend: An Iterative Alignment Framework for Temporal Sentence Grounding ...
|
|
|
|
BASE
|
|
Show details
|
|
97 |
Knowledge Enhanced Fine-Tuning for Better Handling Unseen Entities in Dialogue Generation ...
|
|
|
|
BASE
|
|
Show details
|
|
98 |
STANKER: Stacking Network based on Level-grained Attention-masked BERT for Rumor Detection on Social Media ...
|
|
|
|
BASE
|
|
Show details
|
|
99 |
IndoNLG: Benchmark and Resources for Evaluating Indonesian Natural Language Generation ...
|
|
|
|
BASE
|
|
Show details
|
|
100 |
SYSML: StYlometry with Structure and Multitask Learning: Implications for Darknet Forum Migrant Analysis ...
|
|
|
|
BASE
|
|
Show details
|
|
Page: 1 2 3 4 5 6 7 8 9... 52
|
|