Sep 22, 2023 · In this paper, we propose a novel method to improve machine translation generation by exploiting the source-visual-target consistency mechanism.
Sep 26, 2023 · Experimental results showed that the proposed approach significantly improved translation performance compared to strong baselines for the ...
Dec 27, 2019 · The proposed approach jointly trains the source-to-target and target-to-source translation models and encourages them to share the same focus on ...
Missing: Glancing Enhance
To make better use of visual information, this work presents visual agreement regularized training. The proposed approach jointly trains the source-to- target ...
Missing: Glancing Enhance
Apr 4, 2021 · Glancing Text and Vision Regularized Training to Enhance Machine Translation ... machine translation by jointly learning to align and translate.
Oct 22, 2024 · To sum up, this paper studies the visual-text semantic interaction on the encoder side and the visual-text semantic interaction on the decoder ...
Glancing Text and Vision Regularized Training to Enhance Machine Translation ; Pei Cheng. Zhengzhou University of Light Industry, Zhengzhou, China ; Xiayang Shi.
Experiments on two datasets demonstrate that our approach can effectively enhance the visual awareness of MMT model and achieve superior results against strong ...
Missing: Glancing | Show results with:Glancing
Sep 12, 2023 · In this paper, we propose a novel method that glances future in curriculum learning to achieve the transition from the seq2seq training to prefix2prefix ...
Missing: Regularized | Show results with:Regularized
Sep 1, 2024 · This paper introduces the Vision-guided Target-side Future Context (VisTFC) learning framework for NMT. Our core objective is to refine translation quality.
Missing: Glancing | Show results with:Glancing