Steering Output Style and Topic in Neural Response Generation

By Di Wang, Nebojsa Jojic, Chris Brockett, Eric Nyberg. 2017

68 citations recorded by the source corpus.

We propose simple and flexible training and decoding methods for influencing output style and topic in neural encoder-decoder based language generation. This capability is desirable in a variety of applications, including conversational systems, where successful agents need to produce language in a specific style and generate responses steered by a human puppeteer or external knowledge. We decompose the neural generation process into empirically easier sub-problems: a faithfulness model and a decoding method based on selective-sampling. We also describe training and sampling algorithms that bias the generation process with a specific language style restriction, or a topic restriction. Human evaluation results show that our proposed methods are able to to restrict style and topic without degrading output quality in conversational tasks.

Subfields: Artificial Intelligence, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

Citations in the corpus

Towards a Human-like Open-Domain Chatbot (2020, methodology): “…approach to mitigating this problem is to use more sophisticated decoding algorithms, for instance with different forms of re-ranking (Li et al., 2016a; Shao et al., 2017) or conditioning on proﬁles, topics, and styles (Li et al., 2016b; Wang et al., 2017; Xing et al., 2017; Zhang et al., 2018b).”
What makes a good conversation? How controllable attributes affect human judgments (2019, background): “Controllable neural text generation Researchers have proposed several approaches to control aspects of RNN-based natural language generation such as sentiment, length, speaker style and tense (Fan et al., 2018; Ficler and Goldberg, 2017; Ghazvininejad et al., 2017; Hu et al., 2017; Kikuchi et al., 2”
Adversarial Example Generation with Syntactically Controlled Paraphrase Networks (2018, methodology): “For dialogue, Li et al. (2016a) affect the output using speaker identity, while Wang et al. (2017) develop models to inﬂuence topic and style of the output.”
Towards Controllable Story Generation (2018, background): “With the recent successes on controllable generation of images (Chen et al., 2016; Siddharth et al., 2017; Lample et al., 2017), dialog responses (Wang et al., 2017), poems (Ghazvininejad et al., 2017), and different styles of text (Hu et al., 2017; Ficler and Goldberg, 2017; Shen et al., 2017; Fu…”
Challenges in Building Intelligent Open-domain Dialog Systems (2019, methodology): “[143] utilized a small-scale stylistic data and proposed a topic embedding model to generate responses in specific styles and topics simultaneously.”
Polite Dialogue Generation Without Parallel Data (2018, background): “Clear instructions (closely following those from Wang et al. (2017)) corresponding to each score were shown in the interface.”
Neural personalized response generation as domain adaptation (2017, methodology): “[40] proposed to use small scale style data and a topic embedding model to restrict the style and topic of generated responses.”
Diversifying Dialogue Generation with Non-Conversational Text (2020, background): “Both models share the decoder parameters to facilitate each other.”
Better Conversations by Modeling, Filtering, and Optimizing for Coherence and Diversity (2018, background): “Li et al. (2016b) focus speciﬁcally on personality generation (using personality embeddings) and Wang et al. (2017) promote topic-speciﬁc outputs by language-model rescoring and sampling.”
DialGuide: Aligning Dialogue Model Behavior with Developer Guidelines (2022, background): “, 2020a), steer generation toward a certain topic (Wang et al., 2017; Tang et al., 2019; Gupta et al., 2022a), keywords (Ramakrishnan et al.”
I like fish, especially dolphins: Addressing Contradictions in Dialogue Modeling (2020, background): “The modeling can be implicit where the dialogue consistency-related information like style (Wang et al., 2017), topics, or personal facts are maintained in distributed embeddings (Li et al., 2016; Zhang et al., 2019a), neural long-term memories (Bang et al., 2015), hierarchical neural architecture…”