ScholarGate
Msaidizi
Machine learningDeep learning / NLP / CV

Ufupishaji wa Maandishi Nusu-Simamiwa

Ufupishaji wa maandishi nusu-simamiwa hufunza mifumo ya ufupishaji kwa kutumia kiasi kikubwa cha maandishi yasiyo na lebo pamoja na seti ndogo ya muhtasari wa marejeo ulioandikwa na binadamu. Kwa kutumia mbinu kama vile mafunzo ya awali ya lugha (language-model pretraining), uwekaji lebo bandia (pseudo-labeling), na kujifunza-binafsi (self-training), mbinu hizi hupunguza kwa kiasi kikubwa mzigo wa uwekaji lebo huku zikidumisha alama za ROUGE zenye ushindani kwenye seti za data za vigezo.

Fungua katika MethodMindHivi karibuniVideoHivi karibuniDownload slides

Soma mbinu kamili

Kwa wanachama pekee

Ingia kwa akaunti ya bure ili kusoma sehemu hii.

Ingia

Vyanzo

  1. He, J., Zhou, C., Ma, X., Berg-Kirkpatrick, T., & Neubig, G. (2020). Revisiting Semi-Supervised Learning for Neural Sequence Generation. In Proceedings of ICLR 2020. link
  2. Automatic summarization. Wikipedia. link

Jinsi ya kunukuu ukurasa huu

ScholarGate. (2026, June 3). Semi-supervised Text Summarization (Label-efficient Abstractive and Extractive Summarization). ScholarGate. https://scholargate.app/sw/deep-learning/semi-supervised-text-summarization

ScholarGateSemi-supervised Text Summarization (Semi-supervised Text Summarization (Label-efficient Abstractive and Extractive Summarization)). Imepatikana 2026-06-15 kutoka https://scholargate.app/sw/deep-learning/semi-supervised-text-summarization · Seti ya data: https://doi.org/10.5281/zenodo.20539026