Page: 1 2 3 4 5 6 7 8... 60
61 |
Towards Training Stronger Video Vision Transformers for EPIC-KITCHENS-100 Action Recognition ...
|
|
|
|
BASE
|
|
Show details
|
|
62 |
LICHEE: Improving Language Model Pre-training with Multi-grained Tokenization ...
|
|
|
|
BASE
|
|
Show details
|
|
63 |
Query-graph with Cross-gating Attention Model for Text-to-Audio Grounding ...
|
|
|
|
BASE
|
|
Show details
|
|
64 |
Congolese Swahili Machine Translation for Humanitarian Response ...
|
|
|
|
BASE
|
|
Show details
|
|
65 |
LAWDR: Language-Agnostic Weighted Document Representations from Pre-trained Models ...
|
|
|
|
BASE
|
|
Show details
|
|
66 |
FST: the FAIR Speech Translation System for the IWSLT21 Multilingual Shared Task ...
|
|
|
|
BASE
|
|
Show details
|
|
67 |
Pay Better Attention to Attention: Head Selection in Multilingual and Multi-Domain Sequence Modeling ...
|
|
|
|
BASE
|
|
Show details
|
|
68 |
Code Generation from Natural Language with Less Prior Knowledge and More Monolingual Data ...
|
|
|
|
BASE
|
|
Show details
|
|
69 |
UserAdapter: Few-Shot User Learning in Sentiment Analysis ...
|
|
|
|
BASE
|
|
Show details
|
|
70 |
CLICKER: A Computational LInguistics Classification Scheme for Educational Resources ...
|
|
|
|
BASE
|
|
Show details
|
|
71 |
What Truly Matters? Using Linguistic Cues for Analyzing the #BlackLivesMatter Movement and its Counter Protests: 2013 to 2020 ...
|
|
|
|
BASE
|
|
Show details
|
|
72 |
LAVT: Language-Aware Vision Transformer for Referring Image Segmentation ...
|
|
|
|
BASE
|
|
Show details
|
|
73 |
P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks ...
|
|
|
|
BASE
|
|
Show details
|
|
74 |
CONFIT: Toward Faithful Dialogue Summarization with Linguistically-Informed Contrastive Fine-tuning ...
|
|
Tang, Xiangru; Nair, Arjun; Wang, Borui; Wang, Bingyao; Desai, Jai; Wade, Aaron; Li, Haoran; Celikyilmaz, Asli; Mehdad, Yashar; Radev, Dragomir. - : arXiv, 2021
|
|
Abstract:
Factual inconsistencies in generated summaries severely limit the practical applications of abstractive dialogue summarization. Although significant progress has been achieved by using pre-trained models, substantial amounts of hallucinated content are found during the human evaluation. Pre-trained models are most commonly fine-tuned with cross-entropy loss for text summarization, which may not be an optimal strategy. In this work, we provide a typology of factual errors with annotation data to highlight the types of errors and move away from a binary understanding of factuality. We further propose a training strategy that improves the factual consistency and overall quality of summaries via a novel contrastive fine-tuning, called ConFiT. Based on our linguistically-informed typology of errors, we design different modular objectives that each target a specific type. Specifically, we utilize hard negative samples with errors to reduce the generation of factual inconsistency. In order to capture the key ...
|
|
Keyword:
Computation and Language cs.CL; FOS Computer and information sciences
|
|
URL: https://dx.doi.org/10.48550/arxiv.2112.08713 https://arxiv.org/abs/2112.08713
|
|
BASE
|
|
Hide details
|
|
75 |
Searching for Legal Documents at Paragraph Level: Automating Label Generation and Use of an Extended Attention Mask for Boosting Neural Models of Semantic Similarity ...
|
|
|
|
BASE
|
|
Show details
|
|
76 |
Alpha at SemEval-2021 Tasks 6: Transformer Based Propaganda Classification ...
|
|
|
|
BASE
|
|
Show details
|
|
77 |
The Authors Matter: Understanding and Mitigating Implicit Bias in Deep Text Classification ...
|
|
|
|
BASE
|
|
Show details
|
|
80 |
From Discourse to Narrative: Knowledge Projection for Event Relation Extraction ...
|
|
|
|
BASE
|
|
Show details
|
|
Page: 1 2 3 4 5 6 7 8... 60
|
|