PAL: Persona-Augmented Emotional Support Conversation Generation

Cheng, Jiale; Sabour, Sahand; Sun, Han‐Dong; Chen, Zhuang; Huang, Minlie

doi:10.48550/arxiv.2212.09235

Cited by 2 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After training our persona extractor PESS, we freeze PESS and combine it with response generation model. We refer to this combined model as PESS-GEN. We adopt the architecture proposed in PAL [6] for our response generation model. Figure 2 illustrates the structure of PESS-GEN.…”

Section: Response Generationmentioning

confidence: 99%

“…In response generation evaluation, we compare our model with Blenderbot-Joint [13], MISC [14], GLHG [15], PAL [6] and SUP-PORTER [16]. Note that PAL has the same structure as our model PESS-GEN, except for the persona extractor.…”

Section: Setupmentioning

confidence: 99%

“…We fix τ to 0.9 for all of our experiments. When we train response generation model, we follow setting of [6].…”

Section: Setupmentioning

confidence: 99%

“…This information is called the user's persona. Numerous studies [5,6] have demonstrated the effectiveness of incorporating user persona into dialogue generation. A common approach in these studies has been to rely on pre-provided user persona, rather than inferring the persona during the conversation.…”

Section: Introductionmentioning

confidence: 99%

“…Additionally, the user's persona may be expressed across multiple utterances, requiring a certain level of reasoning. Another approach [6] employed BART [8] which is fine-tuned on the Persona-Chat dataset [9]. Given the user's entire utterances in dialogue history, their model is trained to infer the user's persona which consists of multiple persona sentences.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

School Characteristics and School Violence

Han¹

2021

School Violence in South Korea

View full text Add to dashboard Cite

Providing emotional support through dialogue systems is becoming increasingly important in today's world, as it can support both mental health and social interactions in many conversation scenarios. Previous works have shown that using persona is effective for generating empathetic and supportive responses. They have often relied on pre-provided persona rather than inferring them during conversations. However, it is not always possible to obtain a user persona before the conversation begins. To address this challenge, we propose PESS (Persona Extraction through Semantic Similarity), a novel framework that can automatically infer informative and consistent persona from dialogues. We devise completeness loss and consistency loss based on semantic similarity scores. The completeness loss encourages the model to generate missing persona information, and the consistency loss guides the model to distinguish between consistent and inconsistent persona. Our experimental results demonstrate that high-quality persona information inferred by PESS is effective in generating emotionally supportive responses.

show abstract

Section: Response Generationmentioning

confidence: 99%

Section: Setupmentioning

confidence: 99%

“…We fix τ to 0.9 for all of our experiments. When we train response generation model, we follow setting of [6].…”

Section: Setupmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

School Characteristics and School Violence

Han¹

2021

School Violence in South Korea

View full text Add to dashboard Cite

show abstract

A chatbot for mental health support: exploring the impact of Emohaa on reducing mental distress in China

Sabour

Zhang

Xiao³

et al. 2023

Front. Digit. Health

Self Cite

View full text Add to dashboard Cite

IntroductionThe growing demand for mental health support has highlighted the importance of conversational agents as human supporters worldwide and in China. These agents could increase availability and reduce the relative costs of mental health support. The provided support can be divided into two main types: cognitive and emotional. Existing work on this topic mainly focuses on constructing agents that adopt Cognitive Behavioral Therapy (CBT) principles. Such agents operate based on pre-defined templates and exercises to provide cognitive support. However, research on emotional support using such agents is limited. In addition, most of the constructed agents operate in English, highlighting the importance of conducting such studies in China. To this end, we introduce Emohaa, a conversational agent that provides cognitive support through CBT-Bot exercises and guided conversations. It also emotionally supports users through ES-Bot, enabling them to vent their emotional problems. In this study, we analyze the effectiveness of Emohaa in reducing symptoms of mental distress.Methods and ResultsFollowing the RCT design, the current study randomly assigned participants into three groups: Emohaa (CBT-Bot), Emohaa (Full), and control. With both Intention-To-Treat (N=247) and PerProtocol (N=134) analyses, the results demonstrated that compared to the control group, participants who used two types of Emohaa experienced considerably more significant improvements in symptoms of mental distress, including depression (F[2,244]=6.26, p=0.002), negative affect (F[2,244]=6.09, p=0.003), and insomnia (F[2,244]=3.69, p=0.026).DiscussionBased on the obtained results and participants’ satisfaction with the platform, we concluded that Emohaa is a practical and effective tool for reducing mental distress.

show abstract

PAL: Persona-Augmented Emotional Support Conversation Generation

Cited by 2 publications

References 0 publications

School Characteristics and School Violence

School Characteristics and School Violence

A chatbot for mental health support: exploring the impact of Emohaa on reducing mental distress in China

Contact Info

Product

Resources

About