Image Guided Story Ending Generation On Vist
Metrics
BLEU-1
BLEU-2
BLEU-3
BLEU-4
CIDEr
METEOR
ROUGE-L
Results
Performance results of various models on this benchmark
Model Name | BLEU-1 | BLEU-2 | BLEU-3 | BLEU-4 | CIDEr | METEOR | ROUGE-L | Paper Title | Repository |
|---|---|---|---|---|---|---|---|---|---|
| IE+MSA | 19.15 | 5.74 | 2.73 | 1.63 | 15.56 | 6.59 | 20.62 | Story Ending Generation with Incremental Encoding and Commonsense Knowledge | |
| MGCL | 22.57 | 8.16 | 4.23 | 2.49 | 21.64 | 7.84 | 21.66 | - | - |
| T-CVAE | 14.34 | 5.06 | 2.01 | 1.13 | 11.49 | 4.23 | 15.51 | Hierarchical Photo-Scene Encoder for Album Storytelling | - |
| Seq2Seq | 13.96 | 5.57 | 2.94 | 1.69 | 12.04 | 4.54 | 16.84 | Effective Approaches to Attention-based Neural Machine Translation | |
| Transformer | 17.18 | 6.29 | 3.07 | 2.01 | 12.75 | 6.91 | 18.23 | Attention Is All You Need | |
| MMT | 22.87 | 8.68 | 4.38 | 2.61 | 25.41 | 15.55 | 23.61 | MMT: Image-guided Story Ending Generation with Multimodal Memory Transformer |
0 of 6 row(s) selected.