2 |
Navigating the Kaleidoscope of COVID-19 Misinformation Using Deep Learning ...
|
|
|
|
BASE
|
|
Show details
|
|
3 |
HittER: Hierarchical Transformers for Knowledge Graph Embeddings ...
|
|
|
|
BASE
|
|
Show details
|
|
5 |
HETFORMER: Heterogeneous Transformer with Sparse Attention for Long-Text Extractive Summarization ...
|
|
|
|
BASE
|
|
Show details
|
|
6 |
Not All Negatives are Equal: Label-Aware Contrastive Loss for Fine-grained Text Classification ...
|
|
|
|
BASE
|
|
Show details
|
|
8 |
Unsupervised Multi-View Post-OCR Error Correction With Language Models ...
|
|
|
|
BASE
|
|
Show details
|
|
9 |
AttentionRank: Unsupervised Keyphrase Extraction using Self and Cross Attentions ...
|
|
|
|
BASE
|
|
Show details
|
|
10 |
Automatic Fact-Checking with Document-level Annotations using BERT and Multiple Instance Learning ...
|
|
|
|
BASE
|
|
Show details
|
|
11 |
Towards the Early Detection of Child Predators in Chat Rooms: A BERT-based Approach ...
|
|
|
|
BASE
|
|
Show details
|
|
12 |
Semantic Categorization of Social Knowledge for Commonsense Question Answering ...
|
|
|
|
BASE
|
|
Show details
|
|
13 |
Pre-train or Annotate? Domain Adaptation with a Constrained Budget ...
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Stepmothers are mean and academics are pretentious: What do pretrained language models learn about you? ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
CLIFF: Contrastive Learning for Improving Faithfulness and Factuality in Abstractive Summarization ...
|
|
|
|
Abstract:
Anthology paper link: https://aclanthology.org/2021.emnlp-main.532/ Abstract: We study generating abstractive summaries that are faithful and factually consistent with the given articles. A novel contrastive learning formulation is presented, which leverages both reference summaries, as positive training data, and automatically generated erroneous summaries, as negative training data, to train summarization systems that are better at distinguishing between them. We further design four types of strategies for creating negative samples, to resemble errors made commonly by two state-of-the-art models, BART and PEGASUS, found in our new human annotations of summary errors. Experiments on XSum and CNN/Daily Mail show that our contrastive learning framework is robust across datasets and models. It consistently produces more factual summaries than strong comparisons with post error correction, entailment-based reranking, and unlikelihood training, according to QA-based factuality evaluation. Human judges echo the ...
|
|
Keyword:
Computational Linguistics; Language Models; Machine Learning; Machine Learning and Data Mining; Natural Language Processing; Text Summarization
|
|
URL: https://dx.doi.org/10.48448/zqr1-d338 https://underline.io/lecture/38001-cliff-contrastive-learning-for-improving-faithfulness-and-factuality-in-abstractive-summarization
|
|
BASE
|
|
Hide details
|
|
16 |
Automatic Text Evaluation through the Lens of Wasserstein Barycenters ...
|
|
|
|
BASE
|
|
Show details
|
|
17 |
Combining sentence and table evidence to predict veracity of factual claims using TaPaS and RoBERTa ...
|
|
|
|
BASE
|
|
Show details
|
|
18 |
Meta Distant Transfer Learning for Pre-trained Language Models ...
|
|
|
|
BASE
|
|
Show details
|
|
20 |
Temporal Adaptation of BERT and Performance on Downstream Document Classification: Insights from Social Media ...
|
|
|
|
BASE
|
|
Show details
|
|
|
|