GrounDial: Human-norm Grounded Safe Dialog Response Generation
Authors:
Siwon Kim,
Shuyang Dai,
Mohammad Kachuee,
Shayan Ray,
Tara Taghavi,
Sungroh Yoon
Abstract:
Current conversational AI systems based on large language models (LLMs) are known to generate unsafe responses, agreeing to offensive user input or including toxic content. Previous research aimed to alleviate the toxicity, by fine-tuning LLM with manually annotated safe dialogue histories. However, the dependency on additional tuning requires substantial costs. To remove the dependency, we propos…
▽ More
Current conversational AI systems based on large language models (LLMs) are known to generate unsafe responses, agreeing to offensive user input or including toxic content. Previous research aimed to alleviate the toxicity, by fine-tuning LLM with manually annotated safe dialogue histories. However, the dependency on additional tuning requires substantial costs. To remove the dependency, we propose GrounDial, where response safety is achieved by grounding responses to commonsense social rules without requiring fine-tuning. A hybrid approach of in-context learning and human-norm-guided decoding of GrounDial enables the response to be quantitatively and qualitatively safer even without additional data or tuning.
△ Less
Submitted 14 February, 2024;
originally announced February 2024.
Learning Personalized Representations using Graph Convolutional Network
Authors:
Hongyu Shen,
**oh Oh,
Shuai Zhao,
Guoyin Wang,
Tara Taghavi,
Sung** Lee
Abstract:
Generating representations that precisely reflect customers' behavior is an important task for providing personalized skill routing experience in Alexa. Currently, Dynamic Routing (DR) team, which is responsible for routing Alexa traffic to providers or skills, relies on two features to be served as personal signals: absolute traffic count and normalized traffic count of every skill usage per cust…
▽ More
Generating representations that precisely reflect customers' behavior is an important task for providing personalized skill routing experience in Alexa. Currently, Dynamic Routing (DR) team, which is responsible for routing Alexa traffic to providers or skills, relies on two features to be served as personal signals: absolute traffic count and normalized traffic count of every skill usage per customer. Neither of them considers the network based structure for interactions between customers and skills, which contain richer information for customer preferences. In this work, we first build a heterogeneous edge attributed graph based customers' past interactions with the invoked skills, in which the user requests (utterances) are modeled as edges. Then we propose a graph convolutional network(GCN) based model, namely Personalized Dynamic Routing Feature Encoder(PDRFE), that generates personalized customer representations learned from the built graph. Compared with existing models, PDRFE is able to further capture contextual information in the graph convolutional function. The performance of our proposed model is evaluated by a downstream task, defect prediction, that predicts the defect label from the learned embeddings of customers and their triggered skills. We observe up to 41% improvements on the cross entropy metric for our proposed models compared to the baselines.
△ Less
Submitted 28 July, 2022;
originally announced July 2022.
