H
Huang He
Researcher at Baidu
Publications - 30
Citations - 418
Huang He is an academic researcher from Baidu. The author has contributed to research in topics: Computer science & Conversation. The author has an hindex of 7, co-authored 25 publications receiving 177 citations.
Papers
More filters
Proceedings ArticleDOI
PLATO: Pre-trained Dialogue Generation Model with Discrete Latent Variable
TL;DR: The authors propose a dialogue generation pre-training framework to support various kinds of conversations, including chit-chat, knowledge grounded dialogues, and conversational question answering, which adopts flexible attention mechanisms to fully leverage the bi-directional context and the uni-irectional characteristic of language generation.
Posted Content
PLATO: Pre-trained Dialogue Generation Model with Discrete Latent Variable
TL;DR: This work proposes a novel dialogue generation pre-training framework to support various kinds of conversations, including chit-chat, knowledge grounded dialogues, and conversational question answering, and introduces discrete latent variables to tackle the inherent one-to-many mapping problem in response generation.
Posted Content
PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning
TL;DR: To build a high-quality open-domain chatbot, this work introduces the effective training process of PLATO-2 via curriculum learning, achieving new state-of-the-art results.
Proceedings ArticleDOI
PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning
TL;DR: In this paper, a coarse-grained generation model is trained to learn response generation under the simplified framework of one-to-one mapping and a finegrained generative model augmented with latent variables and an evaluation model are further trained to generate diverse responses and to select the best response, respectively.
Proceedings ArticleDOI
Know More about Each Other: Evolving Dialogue Strategy via Compound Assessment.
TL;DR: In this paper, a generation-evaluation framework is developed for multi-turn conversations with the objective of letting both participants know more about each other, and a dialogue strategy which controls knowledge selection is instantiated and continuously adapted via reinforcement learning.