AI Paper Atlas

Research map / Image generation / Editing and style transfer

Stylized Dialogue Generation with Multi-Pass Dual Learning

Jinpeng Li, Yingce Xia, Rui Yan, Hongda Sun, Dongyan Zhao, Tie-Yan Liu

NeurIPS 2021 · Image generation › Editing and style transfer · approach: Arbitrary neural style transfer · 13 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add quality discriminators that weight pseudo data (ablation)MPDL · TCFC avg of S0 and S1 responsesBLEU-1: 10.9 -> 12.0Improved
add quality discriminators that weight pseudo data (ablation)MPDL · TCFC avg of S0 and S1 responsesDistinct-2: 52.8 -> 50.1Hurt
add quality discriminators that weight pseudo data (ablation)MPDL · TCFC avg of S0 and S1 responsesBERT style accuracy: 83.1 -> 84.4Improved
add multi-pass dual loss L_tra closed loop (ablation)MPDL · TCFC avg of S0 and S1 responsesBLEU-1: 9.89 -> 12.0Improved
add multi-pass dual loss L_tra closed loop (ablation)MPDL · TCFC avg of S0 and S1 responsesDistinct-2: 45.1 -> 50.1Improved
initialize from pretrained DialoGPT (ablation)MPDL · TCFC avg of S0 and S1 responsesBLEU-2: 0.87 -> 1.38Improved
initialize from pretrained DialoGPT (ablation)MPDL · TCFC avg of S0 and S1 responsesDistinct-2: 33.6 -> 50.1Improved
initialize from pretrained DialoGPT (ablation)MPDL · TCFC avg of S0 and S1 responsesBERT style accuracy: 84.0 -> 84.4Mixed
MPDL vs SRJT on formal response generationSRJT · TCFC formal response (S1)BLEU-1: 15.1 -> 16.5Improved
MPDL vs SRJT on Shakespearean response generationSRJT · SDGC Shakespearean response (S1)BLEU-1: 12.2 -> 13.9Improved
MPDL vs SRJT on Shakespearean response generationSRJT · SDGC Shakespearean response (S1)BERT style accuracy: 65.0 -> 64.3Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy