Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
80 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
7 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Structuring Latent Spaces for Stylized Response Generation (1909.05361v1)

Published 3 Sep 2019 in cs.CL and cs.AI

Abstract: Generating responses in a targeted style is a useful yet challenging task, especially in the absence of parallel data. With limited data, existing methods tend to generate responses that are either less stylized or less context-relevant. We propose StyleFusion, which bridges conversation modeling and non-parallel style transfer by sharing a structured latent space. This structure allows the system to generate stylized relevant responses by sampling in the neighborhood of the conversation model prediction, and continuously control the style level. We demonstrate this method using dialogues from Reddit data and two sets of sentences with distinct styles (arXiv and Sherlock Holmes novels). Automatic and human evaluation show that, without sacrificing appropriateness, the system generates responses of the targeted style and outperforms competitive baselines.

User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (7)
  1. Xiang Gao (210 papers)
  2. Yizhe Zhang (127 papers)
  3. Sungjin Lee (46 papers)
  4. Michel Galley (50 papers)
  5. Chris Brockett (37 papers)
  6. Jianfeng Gao (344 papers)
  7. Bill Dolan (45 papers)
Citations (48)