SOTAVerified

Abstractive Text Summarization

Abstractive Text Summarization is the task of generating a short and concise summary that captures the salient ideas of the source text. The generated summaries potentially contain new phrases and sentences that may not appear in the source text.

Source: Generative Adversarial Network for Abstractive Text Summarization

Image credit: Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 51–100 of 846 papers

TitleStatusHype
Leveraging Locality in Abstractive Text SummarizationCode1
FactPEGASUS: Factuality-Aware Pre-training and Fine-tuning for Abstractive SummarizationCode1
ViT5: Pretrained Text-to-Text Transformer for Vietnamese Language GenerationCode1
Masked Summarization to Generate Factually Inconsistent Summaries for Improved Factual Consistency CheckingCode1
Towards Making the Most of Cross-Lingual Transfer for Zero-Shot Neural Machine TranslationCode1
FactGraph: Evaluating Factuality in Summarization with Semantic Graph RepresentationsCode1
Speculative Decoding: Exploiting Speculative Execution for Accelerating Seq2seq GenerationCode1
Towards Abstractive Grounded Summarization of Podcast TranscriptsCode1
ODE Transformer: An Ordinary Differential Equation-Inspired Model for Sequence GenerationCode1
SummaReranker: A Multi-Task Mixture-of-Experts Re-ranking Framework for Abstractive SummarizationCode1
Unsupervised Multi-Granularity SummarizationCode1
CrossSum: Beyond English-Centric Cross-Lingual Summarization for 1,500+ Language PairsCode1
Learning Rich Representation of Keyphrases from TextCode1
LongT5: Efficient Text-To-Text Transformer for Long SequencesCode1
Controlling Conditional Language Models without Catastrophic ForgettingCode1
GupShup: Summarizing Open-Domain Code-Switched ConversationsCode1
Summ^N: A Multi-Stage Summarization Framework for Long Input Dialogues and DocumentsCode1
Towards Making the Most of Multilingual Pretraining for Zero-Shot Neural Machine TranslationCode1
PRIMERA: Pyramid-based Masked Sentence Pre-training for Multi-document SummarizationCode1
Control Prefixes for Parameter-Efficient Text GenerationCode1
DYLE: Dynamic Latent Extraction for Abstractive Long-Input SummarizationCode1
HydraSum: Disentangling Stylistic Features in Text Summarization using Multi-Decoder ModelsCode1
HowSumm: A Multi-Document Summarization Dataset Derived from WikiHow ArticlesCode1
Target-Side Data Augmentation for Sequence GenerationCode1
BARTpho: Pre-trained Sequence-to-Sequence Models for VietnameseCode1
TVStoryGen: A Dataset for Generating Stories with Character DescriptionsCode1
ARMAN: Pre-training with Semantically Selecting and Reordering of Sentences for Persian Abstractive SummarizationCode1
Sequence Level Contrastive Learning for Text SummarizationCode1
Vision Guided Generative Pre-trained Language Models for Multimodal Abstractive SummarizationCode1
Faithful or Extractive? On Mitigating the Faithfulness-Abstractiveness Trade-off in Abstractive SummarizationCode1
Factual Consistency Evaluation for Text Summarization via Counterfactual EstimationCode1
Hallucinated but Factual! Inspecting the Factuality of Hallucinations in Abstractive SummarizationCode1
Enhanced Seq2Seq Autoencoder via Contrastive Learning for Abstractive Text SummarizationCode1
Alleviating Exposure Bias via Contrastive Learning for Abstractive Text SummarizationCode1
EmailSum: Abstractive Email Thread SummarizationCode1
R-Drop: Regularized Dropout for Neural NetworksCode1
XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 LanguagesCode1
Neural Abstractive Unsupervised Summarization of Online News DiscussionsCode1
Attention Temperature Matters in Abstractive Summarization DistillationCode1
SimCLS: A Simple Framework for Contrastive Learning of Abstractive SummarizationCode1
Dissecting Generation Modes for Abstractive Summarization Models via Ablation and AttributionCode1
Enriching Transformers with Structured Tensor-Product Representations for Abstractive SummarizationCode1
ConvoSumm: Conversation Summarization Benchmark and Improved Abstractive Summarization with Argument MiningCode1
Cross-Lingual Abstractive Summarization with Limited Parallel ResourcesCode1
BookSum: A Collection of Datasets for Long-form Narrative SummarizationCode1
The Summary Loop: Learning to Write Abstractive Summaries Without ExamplesCode1
Improving Factual Consistency of Abstractive Summarization via Question AnsweringCode1
Long-Span Summarization via Local Attention and Content SelectionCode1
The Factual Inconsistency Problem in Abstractive Text Summarization: A SurveyCode1
Understanding Factuality in Abstractive Summarization with FRANK: A Benchmark for Factuality MetricsCode1
Show:102550
← PrevPage 2 of 17Next →
All datasetsCNN / Daily MailAbstractive Text Summarization from Il PostvietnewsAbstractive Text Summarization from FanpageEDUsumCNN/Daily MailMLSum-itWITSAESLCeLifeInshorts NewsMLSUM de

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Scrambled code + broken (alter)ROUGE-148.18—Unverified
2BRIOROUGE-147.78—Unverified
3PegasusROUGE-147.36—Unverified
4PEGASUS + SummaRerankerROUGE-147.16—Unverified
5Scrambled code + brokenROUGE-146.71—Unverified
6BART + SimCLSROUGE-146.67—Unverified
7SEASONROUGE-146.27—Unverified
8Fourier TransformerROUGE-144.76—Unverified
9GLM-XXLargeROUGE-144.7—Unverified
10BART + R-DropROUGE-144.51—Unverified
#ModelMetricClaimedVerifiedStatus
1mBARTROUGE-138.91—Unverified
2mBARTROUGE-138.91—Unverified
3BART-ITROUGE-137.31—Unverified
4mT5ROUGE-135.04—Unverified
5IT5ROUGE-133.78—Unverified
6IT5-baseROUGE-132.88—Unverified
7Pegasus-CNN/DM (eng-it translation)ROUGE-123.96—Unverified
8Pegasus-XSum (eng-it translation)ROUGE-121.03—Unverified
#ModelMetricClaimedVerifiedStatus
1Kết quả nghiên cứuRouge-167.8—Unverified
2ViT5 largeRouge-163.37—Unverified
3ViT5 baseRouge-162.77—Unverified
4BARTphoRouge-161.14—Unverified
5mBARTRouge-159.81—Unverified
6mT5Rouge-158.05—Unverified
7TransformerRouge-157.56—Unverified
#ModelMetricClaimedVerifiedStatus
1mBARTROUGE-136.52—Unverified
2mBARTROUGE-136.5—Unverified
3BART-ITROUGE-135.42—Unverified
4mT5ROUGE-134.13—Unverified
5IT5-baseROUGE-133.99—Unverified
6IT5ROUGE-133.83—Unverified
#ModelMetricClaimedVerifiedStatus
1GP_Step_SimROUGE-164.48—Unverified
2NEZHAROUGE-163.91—Unverified
3RoBERTaROUGE-163.22—Unverified
4BERTROUGE-162.37—Unverified
5Seq2seqROUGE-148.62—Unverified
#ModelMetricClaimedVerifiedStatus
1BART (TextBox 2.0)ROUGE-144.47—Unverified
2SEGMENTROUGE-142.1—Unverified
3CriSPO 3-shotROUGE-142.1—Unverified
4Claude Instant + SigExtROUGE-142—Unverified
#ModelMetricClaimedVerifiedStatus
1mBARTrouge119.35—Unverified
2IT5rouge119.29—Unverified
3Pegasus-CNN/DM (eng-it translation)rouge116.97—Unverified
4Pegasus-XSum (eng-it translation)rouge115.17—Unverified
#ModelMetricClaimedVerifiedStatus
1BART-ITROUGE-142.32—Unverified
2mT5ROUGE-140.6—Unverified
3mBARTROUGE-139.32—Unverified
4IT5-baseROUGE-137.98—Unverified
#ModelMetricClaimedVerifiedStatus
1PEGASUSROUGE-137.68—Unverified
2Multi-Stage Extractor/AbstractorROUGE-123.67—Unverified
#ModelMetricClaimedVerifiedStatus
1Two stage LLMsTest ROGUE-10.46—Unverified
#ModelMetricClaimedVerifiedStatus
1T2SAMROUGE48.5—Unverified
#ModelMetricClaimedVerifiedStatus
1mBARTMETEOR0.44—Unverified
#ModelMetricClaimedVerifiedStatus
1mBARTMETEOR0.21—Unverified
#ModelMetricClaimedVerifiedStatus
1Two stage LLMsTest ROGUE-10.44—Unverified
#ModelMetricClaimedVerifiedStatus
1BertSumContent F129.8—Unverified