PK PyTorchKR Papers
Papers with Code · CC BY-SA 4.0

Text Summarization

473 평가 기록 92 datasets

Text Summarization is a natural language processing (NLP) task that involves condensing a lengthy text document into a shorter, more compact version while still retaining the most important information and meaning. The goal is to produce a summary that accurately represents the content of the original text in a concise form.

There are different approaches to text summarization, including extractive methods that identify and extract important sentences or phrases from the text, and abstractive methods that generate new text based on the content of the original text.

개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.

벤치마크

Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.

Dataset기록된 모델 예시MetricPaperCode
Scrambled code + broken (alter)
ROUGE-1: 48.18ROUGE-2: 19.84ROUGE-L: 45.35
논문 코드
OpenAI/o3-mini
ROUGE-1: 60.12ROUGE-2: 54.22ROUGE-L: 57.21
- -
Top Down Transformer (AdaPool) (464M)
ROUGE-1: 51.05ROUGE-2: 23.26ROUGE-L: 46.47
논문 코드
Top Down Transformer (AdaPool) (464M)
ROUGE-1: 50.95ROUGE-2: 21.93ROUGE-L: 45.61
논문 코드
Scrambled code + broken (alter)
ROUGE-1: 48.18ROUGE-2: 19.84ROUGE-L: 45.35
논문 코드
MPNet-multilingual
Spearman Correlation: 31.57
논문 코드
Selfmem
ROUGE-1: 50.30ROUGE-2: 26.70ROUGE-3: 41.60
논문 코드
HAHSum
ROUGE-2: 21.30ROUGE-1: 44.68ROUGE-L: 40.75
논문 -
HSSAS
ROUGE-1: 42.3ROUGE-2: 17.8ROUGE-L: 37.6
논문 -
Transformer+WDrop
ROUGE-1: 33.06ROUGE-2: 11.45ROUGE-L: 28.51
논문 코드
OmniVec2
ROUGE-1: 59.1ROUGE-2: 34.1ROUGE-L: 63.7BertScoreF1: 65.1
논문 -
LexRank (query: step title)
ROUGE-1: 39.6
논문 코드

데이터셋

92개
CNN / Daily MailGigaWordPubmedArxiv HEP-TH citation graphCNN / Daily MailMTEBX-SumCNN / Daily MailCNN / Daily Mail (Anonymized)DUC 2004 Task 1SAMSumHowSumm-Step

논문

라이브러리

Papers with Code 리더보드 row에 연결된 구현 저장소입니다.

github.com/basf/chemteb 1 논문
github.com/climsocana/tecb-de 1 논문
github.com/embeddings-benchmark/mteb 1 논문
github.com/HHousen/TransformerSum 1 논문
github.com/lyon-nlp/mteb-french 1 논문
github.com/maszhongming/MatchSum 1 논문
github.com/wadoodabdul/clinical_ner_benchmark 1 논문
github.com/alebryvas/berk266 1 논문

하위 태스크

AT Abstractive Text Summarization 23 데이터셋 · 101 rows DS Document Summarization 9 데이터셋 · 56 rows LN Long-Form Narrative Summarization 4 데이터셋 · 30 rows ET Extractive Text Summarization 5 데이터셋 · 22 rows MS Multi-Document Summarization 5 데이터셋 · 10 rows UO Unsupervised Opinion Summarization 3 데이터셋 · 9 rows SC Sentence Compression 1 데이터셋 · 6 rows SD Scientific Document Summarization 3 데이터셋 · 3 rows QE Query-Based Extractive Summarization 1 데이터셋 · 2 rows SC Summarization Consistency Evaluation 1 데이터셋 · 1 rows OS Opinion Summarization 0 데이터셋 · 0 rows SS Sentence Summarization 0 데이터셋 · 0 rows