CLIP Text Encoder (RL w/ CIDEr-reward)

Reported on 4 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing4 results

Image CaptioningonCOCO Captions
BLEU-4· 2022-05-26
38.2
best: 46.5 (mPLUG)
Fine-grained Image Captioning with CLIP Reward arXiv:2205.13115
Image CaptioningonCOCO Captions
CIDER· 2022-05-26
124.9
best: 155.1 (mPLUG)
Fine-grained Image Captioning with CLIP Reward arXiv:2205.13115
Image CaptioningonCOCO Captions
METEOR· 2022-05-26
28.7
best: 33.9 (CoCa)
Fine-grained Image Captioning with CLIP Reward arXiv:2205.13115
Image CaptioningonCOCO Captions
ROUGE-L· 2022-05-26
58.5
best: 61.1 (ExpansionNet v2 (No VL pretraining))
Fine-grained Image Captioning with CLIP Reward arXiv:2205.13115