25 citations · 25 across the 1 of their papers we have counts for
Showing cs.CVShow all
2 papers · 1 filter
cs.CV2019★ 25 cited
Masked Non-Autoregressive Image Captioning
Junlong Gao, Xi Meng, Shiqi Wang +4
Existing captioning models often adopt the encoder-decoder architecture, where the decoder uses autoregressive decoding to generate captions, such that each token is generated sequ…
cs.CV2019
Self-critical n-step Training for Image Captioning
Junlong Gao, Shiqi Wang, Shanshe Wang +2
Existing methods for image captioning are usually trained by cross entropy loss, which leads to exposure bias and the inconsistency between the optimizing function and evaluation m…