A collection of papers and resources related to Large Language Models.
The organization of papers refers to our survey "Neural Headline Generation: A Comprehensive Survey".
Strategy | Work | Code | Model Description | Learning Framework |
---|---|---|---|---|
Extractive | MLRank[87] | Transformer+ Non-autoregressive | Seq2seq | |
Abstractive | NACC[137] | [Code] | Transformer+ Non-autoregressive | Seq2seq |
NAUS[136] | [Code] | Transformer+ Non-autoregressive | Seq2seq | |
TI-C-NHG[171] | Transformer + Copy Mechanism | Seq2seq | ||
IT5[151] | [Code] | T5 | Transfer Learning | |
Heng[57] | UNILM | Adversarial Learning | ||
Amin et al.[91] | GRU + Attention | Seq2seq | ||
Kanungo et al.[70] | MLM + Transformer | Reinforcement Learning | ||
SLGen[95] | GCN + GRU + Attention | Seq2seq | ||
DeepTitle[6] | BERT | Transfer Learning | ||
HG-News[3] | GPT-2 + Pointer Network | Transfer Learning | ||
Matsushita et al.[94] | BART | Transfer Learning | ||
PENS[30] | [Code] | Transformer + Pointer Network | Seq2seq | |
ZmBART[80] | mBART | Transfer Learning | ||
Shavrina et al.[172] | ruGPT-3 | Fine-tuning | ||
CNHG[161] | BiGRU + Attention | Reinforcement Learning | ||
Shu et al.[130] | Variational Auto-Encoder+RNN | Seq2seq | ||
Littman et al.[144] | BERT | Fine-tuning | ||
Scarlatos et al.[67] | [Code] | GPT-2 | Fine-tuning | |
Jang et al.[170] | Transformer | Adversarial Learning | ||
Hybrid | MuD2H[96] | GCN + BiLSTM | Seq2seq | |
SHEG[5] | GRU + CNN + BiLSTM | Seq2seq | ||
Song et al.[104] | BiLSTM | Reinforcement Learning | ||
Liu et al.[78] | BERT | Transfer Learning |
Methods | Work | Code | Description | Style |
---|---|---|---|---|
Pre-processing | Littman et al.[144] | Construct datasets | Satirical | |
Lorenzo et al.[127] | Train on style datasets | |||
In-process | Zhan et al.[173] | Multi-task learning | ||
Song et al.[104] | Popularity classifier | Attractive | ||
TitleStylist[10] | [Code] | Multi-task learning | Clickbait | |
Li et al.[74] | Style constraints module | Attractive | ||
Shu et al.[130] | Style discriminator | Clickbait | ||
Xu et al.[8] | Reinforce learning | Clickbait | ||
Post-processing | Alnajjar et al.[9] | Word replacement | Humor | |
Stegeren et al.[1] | Word replacement | |||
Alnajjar et al.[77] | Headline replacement | Metaphorical | ||
Fu et al.[193] | Style transfer model |