A Persona-Based Neural Conversation Model

By Jiwei Li, Michel Galley, Chris Brockett, Georgios P. Spithourakis, Jianfeng Gao, William B. Dolan. 2016

1,107 citations recorded by the source corpus.

We present persona-based models for handling the issue of speaker consistency in neural response generation. A speaker model encodes personas in distributed embeddings that capture individual characteristics such as background information and speaking style. A dyadic speaker-addressee model captures properties of interactions between two interlocutors. Our models yield qualitative performance improvements in both perplexity and BLEU scores over baseline sequence-to-sequence models, with similar gains in speaker consistency as measured by human judges.

Subfields: Artificial Intelligence, Natural Language Processing, Human-Computer Interaction

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

In Human-Computer Interaction

Citations in the corpus

Towards a Human-like Open-Domain Chatbot (2020, methodology): “…typically respond with a generic apology when a human asks something outside their domain; some end-to-end learned chatbots respond “I don’t know” to many inputs (Li et al., 2016a); and Turing Test contest entrants often try to avoid detection by being strategically vague (Venkatesh et al., 2018).”
Open-Domain Conversational Agents: Current Progress, Open Problems, and Future Directions (2020, methodology): “In order to solve this two strategies have been tried: trying to learn to model the personality as part of the learned weights given the speaker id from the data (Li et al., 2016), or providing training data with explicit personality information.”
ACUTE-EVAL: Improved Dialogue Evaluation with Optimized Questions and Multi-turn Comparisons (2019, methodology): “The two most used approaches, single-turn pairwise evaluation (Vinyals and Le 2015; Li et al. 2016b), and multi-turn Likert scores (Venkatesh et al. 2017; Zhang et al. 2018; See et al. 2019; Dinan et al. 2019b; Dinan et al. 2019a) have serious limitations.”
Towards Exploiting Background Knowledge for Building Conversation Systems (2018, background): “In the space of academic datasets, Lowe et al. (2015a) report results on the Ubuntu dataset using manpages as external knowledge whereas Ghazvininejad et al.”
Learning from Dialogue after Deployment: Feed Yourself, Chatbot! (2019, background): “Since the question of how to incorporate external knowledge (such as profiles) in dialogue is an open research question of its own (Li et al., 2016; Luan et al., 2017; Luo et al., 2018) and we are primarily interested in the question of learning from dialogue, we discard the profiles and simply trai”
Queens Are Powerful Too: Mitigating Gender Bias in Dialogue Generation (2019, background): “Recent work in dialogue incorporates personas, or personality descriptions that ground speaker’s chat, such as I love fishing (Zhang et al., 2018; Shuster et al., 2018; Mazaré et al., 2018; Olabiyi et al., 2018; Li et al., 2016).”
Towards Coherent and Engaging Spoken Dialog Response Generation Using Automatic Conversation Evaluators (2019, methodology): “We also measured different systems performance using Distinct-2 (Li et al., 2016a), which is the number of unique length-normalized bigrams in responses.”
Towards Coherent and Cohesive Long-form Text Generation (2018, background): “Only recently have there been efforts to provide more relevant objectives, such as consistency and repetition in a text (Li et al., 2015, 2016a; Holtzman et al., 2018).”
Recipes for Building an Open-Domain Chatbot (2020, neutral): “This mirrors results found in some recent papers comparing generation and retrieval (Li et al., 2016; Dinan et al., 2019c).”
DIALOGPT : Large-Scale Generative Pre-training for Conversational Response Generation (2019, methodology): “To address this problem, we implement a maximum mutual information (MMI) scoring function (Li et al., 2016a; Zhang et al., 2018).”
How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation (2016, methodology): “These metrics have recently been adopted by dialogue researchers (Ritter et al., 2011; Sordoni et al., 2015; Li et al., 2015; Galley et al., 2015b; Wen et al., 2015; Li et al., 2016).”
Personalizing Dialogue Agents: I have a dog, do you have pets too? (2018, methodology): “2206 the chit-chat setting, the most relevant work is (Li et al., 2016a).”
The Second Conversational Intelligence Challenge (ConvAI2) (2019, background): “Common issues with chit-chat models include: (i) the lack of a consistent personality [4] as they are typically trained over many dialogues each with diﬀerent speakers, (ii) the lack of an explicit long-term memory as they are typically trained to produce an utterance given only the recent dialogue…”
Adversarial Learning for Neural Dialogue Generation (2017, methodology): “Our work adapts the encoder-decoder model to RL training, and can thus be viewed as an extension of Li et al. (2016d), but with more general RL rewards.”
Deep Reinforcement Learning for Dialogue Generation (2016, methodology): “Human Evaluation We explore three settings for human evaluation: the first setting is similar to what was described in Li et al. (2016a), where we employ crowdsourced judges to evaluate a random sample of 500 items.”
Neural Approaches to Conversational AI (2018, positive): “, 2017) is not a persona-based response generation model per se, their work shares some similarities with speaker embedding models such as (Li et al., 2016b).”
Towards Empathetic Open-domain Conversation Models: A New Benchmark and Dataset (2018, methodology): “(Li et al., 2016b; Zhang et al., 2018; Mazare et al., 2018), focusing on personal facts (“I am from New York”).”