English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The existing image captioning approaches typically train a one-stage sentencedecoder, which is difficult to generate rich fine-grained descriptions. On theother hand, multi-stage image caption model is hard to train due to thevanishing gradient problem. In this paper, we propose a coarse-to-finemulti-stage prediction framework for image captioning, composed of multipledecoders each of which operates on the output of the previous stage, producingincreasingly refined image descriptions. Our proposed learning approachaddresses the difficulty of vanishing gradients during training by providing alearning objective function that enforces intermediate supervisions.Particularly, we optimize our model with a reinforcement learning approachwhich utilizes the output of each intermediate decoder's test-time inferencealgorithm as well as the output of its preceding decoder to normalize therewards, which simultaneously solves the well-known exposure bias problem andthe loss-evaluation mismatch problem. We extensively evaluate the proposedapproach on MSCOCO and show that our approach can achieve the state-of-the-artperformance.

Stack-Captioning: Coarse-to-Fine Learning for Image Captioning