Hierarchically Structured Reinforcement Learning for Topically Coherent Visual Story Generation
Open paper
Benchmark
Model
Rank
Results
visual-storytelling-on-vist
HSRL w/ Joint Training
–
BLEU-4: 12.32
CIDEr: 10.71
METEOR: 35.23
ROUGE-L: 30.84
…
Rank counts only results with a code link.