Pub Date : 2024-02-01Epub Date: 2024-03-24DOI: 10.1609/aaai.v38i21.30326
Bhanu Teja Gullapalli, Stephanie Carreiro, Brittany P Chapman, Eric L Garland, Tauhidur Rahman
Long-term and high-dose prescription opioid use places individuals at risk for opioid misuse, opioid use disorder (OUD), and overdose. Existing methods for monitoring opioid use and detecting misuse rely on self-reports, which are prone to reporting bias, and toxicology testing, which may be infeasible in outpatient settings. Although wearable technologies for monitoring day-to-day health metrics have gained significant traction in recent years due to their ease of use, flexibility, and advancements in sensor technology, their application within the opioid use space remains underexplored. In the current work, we demonstrate that oral opioid administrations can be detected using physiological signals collected from a wrist sensor. More importantly, we show that models informed by opioid pharmacokinetics increase reliability in predicting the timing of opioid administrations. Forty-two individuals who were prescribed opioids as a part of their medical treatment in-hospital and after discharge were enrolled. Participants wore a wrist sensor throughout the study, while opioid administrations were tracked using electronic medical records and self-reports. We collected 1,983 hours of sensor data containing 187 opioid administrations from the inpatient setting and 927 hours of sensor data containing 40 opioid administrations from the outpatient setting. We demonstrate that a self-supervised pre-trained model, capable of learning the canonical time series of plasma concentration of the drug derived from opioid pharmacokinetics, can reliably detect opioid administration in both settings. Our work suggests the potential of pharmacokinetic-informed, data-driven models to objectively detect opioid use in daily life.
{"title":"Pharmacokinetics-Informed Neural Network for Predicting Opioid Administration Moments with Wearable Sensors.","authors":"Bhanu Teja Gullapalli, Stephanie Carreiro, Brittany P Chapman, Eric L Garland, Tauhidur Rahman","doi":"10.1609/aaai.v38i21.30326","DOIUrl":"10.1609/aaai.v38i21.30326","url":null,"abstract":"<p><p>Long-term and high-dose prescription opioid use places individuals at risk for opioid misuse, opioid use disorder (OUD), and overdose. Existing methods for monitoring opioid use and detecting misuse rely on self-reports, which are prone to reporting bias, and toxicology testing, which may be infeasible in outpatient settings. Although wearable technologies for monitoring day-to-day health metrics have gained significant traction in recent years due to their ease of use, flexibility, and advancements in sensor technology, their application within the opioid use space remains underexplored. In the current work, we demonstrate that oral opioid administrations can be detected using physiological signals collected from a wrist sensor. More importantly, we show that models informed by opioid pharmacokinetics increase reliability in predicting the timing of opioid administrations. Forty-two individuals who were prescribed opioids as a part of their medical treatment in-hospital and after discharge were enrolled. Participants wore a wrist sensor throughout the study, while opioid administrations were tracked using electronic medical records and self-reports. We collected 1,983 hours of sensor data containing 187 opioid administrations from the inpatient setting and 927 hours of sensor data containing 40 opioid administrations from the outpatient setting. We demonstrate that a self-supervised pre-trained model, capable of learning the canonical time series of plasma concentration of the drug derived from opioid pharmacokinetics, can reliably detect opioid administration in both settings. Our work suggests the potential of pharmacokinetic-informed, data-driven models to objectively detect opioid use in daily life.</p>","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"38 21","pages":"22892-22898"},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11027727/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140861820","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Utilizing electronic health records (EHR) for machine learning-driven clinical research has great potential to enhance outcome predictions and treatment personalization. Nonetheless, due to privacy and security concerns, the secondary use of EHR data is regulated, constraining researchers' access to EHR data. Generating synthetic EHR data with deep learning methods is a viable and promising approach to mitigate privacy concerns, offering not only a supplementary resource for downstream applications but also sidestepping the privacy risks associated with real patient data. While prior efforts have concentrated on EHR data synthesis, significant challenges persist: addressing the heterogeneity of features including temporal and non-temporal features, structurally missing values, and irregularity of the temporal measures, and ensuring rigorous privacy of the real data used for model training. Existing works in this domain only focused on solving one or two aforementioned challenges. In this work, we propose IGAMT, an innovative framework to generate privacy-preserved synthetic EHR data that not only maintains high quality with heterogeneous features, missing values, and irregular measures but also achieves differential privacy with enhanced privacy-utility trade-off. Extensive experiments prove that IGAMT significantly outperforms baseline and state-of-the-art models in terms of resemblance to real data and performance of downstream applications. Ablation studies also prove the effectiveness of the techniques applied in IGAMT.
{"title":"IGAMT: Privacy-Preserving Electronic Health Record Synthesization with Heterogeneity and Irregularity.","authors":"Wenjie Wang, Pengfei Tang, Jian Lou, Yuanming Shao, Lance Waller, Yi-An Ko, Li Xiong","doi":"10.1609/aaai.v38i14.29491","DOIUrl":"https://doi.org/10.1609/aaai.v38i14.29491","url":null,"abstract":"<p><p>Utilizing electronic health records (EHR) for machine learning-driven clinical research has great potential to enhance outcome predictions and treatment personalization. Nonetheless, due to privacy and security concerns, the secondary use of EHR data is regulated, constraining researchers' access to EHR data. Generating synthetic EHR data with deep learning methods is a viable and promising approach to mitigate privacy concerns, offering not only a supplementary resource for downstream applications but also sidestepping the privacy risks associated with real patient data. While prior efforts have concentrated on EHR data synthesis, significant challenges persist: addressing the heterogeneity of features including temporal and non-temporal features, structurally missing values, and irregularity of the temporal measures, and ensuring rigorous privacy of the real data used for model training. Existing works in this domain only focused on solving one or two aforementioned challenges. In this work, we propose <i>IGAMT</i>, an innovative framework to generate privacy-preserved synthetic EHR data that not only maintains high quality with heterogeneous features, missing values, and irregular measures but also achieves differential privacy with enhanced privacy-utility trade-off. Extensive experiments prove that <i>IGAMT</i> significantly outperforms baseline and state-of-the-art models in terms of resemblance to real data and performance of downstream applications. Ablation studies also prove the effectiveness of the techniques applied in <i>IGAMT</i>.</p>","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"38 14","pages":"15634-15643"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11606572/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142775537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Original Article was published on 26 June 2023.
原文发表于2023年6月26日。
{"title":"Erratum to: 3D-TOGO: Towards Text-Guided Cross-Category 3D Object Generation","authors":"Zutao Jiang, Guansong Lu, Xiaodan Liang, Jihua Zhu, Wei Zhang, Xiaojun Chang, Hang Xu","doi":"10.1609/aaai.v37i13.27320","DOIUrl":"https://doi.org/10.1609/aaai.v37i13.27320","url":null,"abstract":"The Original Article was published on 26 June 2023. \u0000 ","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82809113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-26DOI: 10.1609/aaai.v37i13.26924
Catherine Ordun
Visible-to-Thermal (VT) face translation is an under-studied problem of image-to-image translation that offers an AI-enabled alternative to traditional thermal sensors. Over three phases, my Doctoral Proposal explores developing multimodal deep generative solutions that can be applied towards telemedicine applications. These include the contribution of a novel Thermal Face Contrastive GAN (TFC-GAN), exploration of hybridized diffusion-GAN models, application on real clinical thermal data at the National Institutes of Health, and exploration of strategies for Federated Learning (FL) in heterogenous data settings.
{"title":"Multimodal Deep Generative Models for Remote Medical Applications","authors":"Catherine Ordun","doi":"10.1609/aaai.v37i13.26924","DOIUrl":"https://doi.org/10.1609/aaai.v37i13.26924","url":null,"abstract":"Visible-to-Thermal (VT) face translation is an under-studied problem of image-to-image translation that offers an AI-enabled alternative to traditional thermal sensors. Over three phases, my Doctoral Proposal explores developing multimodal deep generative solutions that can be applied towards telemedicine applications. These include the contribution of a novel Thermal Face Contrastive GAN (TFC-GAN), exploration of hybridized diffusion-GAN models, application on real clinical thermal data at the National Institutes of Health, and exploration of strategies for Federated Learning (FL) in heterogenous data settings.","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"50 1","pages":"16127-16128"},"PeriodicalIF":0.0,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74147410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-26DOI: 10.1609/aaai.v37i5.25813
Daoming Zong, Shiliang Sun
Physical commonsense reasoning is essential for building reliable and interpretable AI systems, which involves a general understanding of the physical properties and affordances of everyday objects, how these objects can be manipulated, and how they interact with others. It is fundamentally a multi-modal task, as physical properties are manifested through multiple modalities, including vision and acoustics. In this work, we present a unified framework, named Multimodal Commonsense Transformer (MCOMET), for physical audiovisual commonsense reasoning. MCOMET has two intriguing properties: i) it fully mines higher-ordered temporal relationships across modalities (e.g., pairs, triplets, and quadruplets); and ii) it restricts the cross-modal flow through the feature collection and propagation mechanism along with tight fusion bottlenecks, forcing the model to attend the most relevant parts in each modality and suppressing the dissemination of noisy information. We evaluate our model on a very recent public benchmark, PACS. Results show that MCOMET significantly outperforms a variety of strong baselines, revealing powerful multi-modal commonsense reasoning capabilities.
{"title":"McOmet: Multimodal Fusion Transformer for Physical Audiovisual Commonsense Reasoning","authors":"Daoming Zong, Shiliang Sun","doi":"10.1609/aaai.v37i5.25813","DOIUrl":"https://doi.org/10.1609/aaai.v37i5.25813","url":null,"abstract":"Physical commonsense reasoning is essential for building reliable and interpretable AI systems, which involves a general understanding of the physical properties and affordances of everyday objects, how these objects can be manipulated, and how they interact with others. It is fundamentally a multi-modal task, as physical properties are manifested through multiple modalities, including vision and acoustics. In this work, we present a unified framework, named Multimodal Commonsense Transformer (MCOMET), for physical audiovisual commonsense reasoning. MCOMET has two intriguing properties: i) it fully mines higher-ordered temporal relationships across modalities (e.g., pairs, triplets, and quadruplets); and ii) it restricts the cross-modal flow through the feature collection and propagation mechanism along with tight fusion bottlenecks, forcing the model to attend the most relevant parts in each modality and suppressing the dissemination of noisy information. We evaluate our model on a very recent public benchmark, PACS. Results show that MCOMET significantly outperforms a variety of strong baselines, revealing powerful multi-modal commonsense reasoning capabilities.","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"42 1","pages":"6621-6629"},"PeriodicalIF":0.0,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75194890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-26DOI: 10.1609/aaai.v37i3.25367
Wu‐ru Wang, Lei Zhang, Hua Huang
Constructing accurate training tuples is crucial for unsupervised local descriptor learning, yet challenging due to the absence of patch labels. The state-of-the-art approach constructs tuples with heuristic rules, which struggle to precisely depict real-world patch transformations, in spite of enabling fast model convergence. A possible solution to alleviate the problem is the clustering-based approach, which can capture realistic patch variations and learn more accurate class decision boundaries, but suffers from slow model convergence. This paper presents HybridDesc, an unsupervised approach that learns powerful local descriptor models with fast convergence speed by combining the rule-based and clustering-based approaches to construct training tuples. In addition, HybridDesc also contributes two concrete enhancing mechanisms: (1) a Differentiable Hyperparameter Search (DHS) strategy to find the optimal hyperparameter setting of the rule-based approach so as to provide accurate prior for the clustering-based approach, (2) an On-Demand Clustering (ODC) method to reduce the clustering overhead of the clustering-based approach without eroding its advantage. Extensive experimental results show that HybridDesc can efficiently learn local descriptors that surpass existing unsupervised local descriptors and even rival competitive supervised ones.
{"title":"Revisiting Unsupervised Local Descriptor Learning","authors":"Wu‐ru Wang, Lei Zhang, Hua Huang","doi":"10.1609/aaai.v37i3.25367","DOIUrl":"https://doi.org/10.1609/aaai.v37i3.25367","url":null,"abstract":"Constructing accurate training tuples is crucial for unsupervised local descriptor learning, yet challenging due to the absence of patch labels. The state-of-the-art approach constructs tuples with heuristic rules, which struggle to precisely depict real-world patch transformations, in spite of enabling fast model convergence. A possible solution to alleviate the problem is the clustering-based approach, which can capture realistic patch variations and learn more accurate class decision boundaries, but suffers from slow model convergence. This paper presents HybridDesc, an unsupervised approach that learns powerful local descriptor models with fast convergence speed by combining the rule-based and clustering-based approaches to construct training tuples. In addition, HybridDesc also contributes two concrete enhancing mechanisms: (1) a Differentiable Hyperparameter Search (DHS) strategy to find the optimal hyperparameter setting of the rule-based approach so as to provide accurate prior for the clustering-based approach, (2) an On-Demand Clustering (ODC) method to reduce the clustering overhead of the clustering-based approach without eroding its advantage. Extensive experimental results show that HybridDesc can efficiently learn local descriptors that surpass existing unsupervised local descriptors and even rival competitive supervised ones.","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"26 1","pages":"2680-2688"},"PeriodicalIF":0.0,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75665780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-26DOI: 10.1609/aaai.v37i13.27077
Di Jia, Qianqian Wang, Jun Cao, Peng Cai, Zhiyang Jin
In this work, we propose a fast convergence track net, or FC-TrackNet, based on a synthetic data-driven approach to maintaining long-term 6D pose tracking. Comparison experiments are performed on two different datasets, The results demonstrate that our approach can achieve a consistent tracking frequency of 90.9 Hz as well as higher accuracy than the state-of-the art approaches.
{"title":"FC-TrackNet: Fast Convergence Net for 6D Pose Tracking in Synthetic Domains","authors":"Di Jia, Qianqian Wang, Jun Cao, Peng Cai, Zhiyang Jin","doi":"10.1609/aaai.v37i13.27077","DOIUrl":"https://doi.org/10.1609/aaai.v37i13.27077","url":null,"abstract":"In this work, we propose a fast convergence track net, or FC-TrackNet, based on a synthetic data-driven approach to maintaining long-term 6D pose tracking. Comparison experiments are performed on two different datasets, The results demonstrate that our approach can achieve a consistent tracking frequency of 90.9 Hz as well as higher accuracy than the state-of-the art approaches.","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"32 1","pages":"16455-16457"},"PeriodicalIF":0.0,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74441212","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-26DOI: 10.1609/aaai.v37i13.26907
Weizhen Bian, Yijin Song, Nianzhen Gu, Tin Yan Chan, Tsz To Lo, Tsun Sun Li, King Chak Wong, Wei Xue, R. Trillo
The significant development of artificial neural network architectures has facilitated the increasing adoption of automated music composition models over the past few years. However, most existing systems feature algorithmic generative structures based on hard code and predefined rules, generally excluding interactive or improvised behaviors. We propose a motion based music system, MoMusic, as a AI real time music generation system. MoMusic features a partially randomized harmonic sequencing model based on a probabilistic analysis of tonal chord progressions, mathematically abstracted through musical set theory. This model is presented against a dual dimension grid that produces resulting sounds through a posture recognition mechanism. A camera captures the users' fingers' movement and trajectories, creating coherent, partially improvised harmonic progressions. MoMusic integrates several timbrical registers, from traditional classical instruments such as the piano to a new ''human voice instrument'' created using a voice conversion technique. Our research demonstrates MoMusic's interactiveness, ability to inspire musicians, and ability to generate coherent musical material with various timbrical registers. MoMusic's capabilities could be easily expanded to incorporate different forms of posture controlled timbrical transformation, rhythmic transformation, dynamic transformation, or even digital sound processing techniques.
{"title":"MoMusic: A Motion-Driven Human-AI Collaborative Music Composition and Performing System","authors":"Weizhen Bian, Yijin Song, Nianzhen Gu, Tin Yan Chan, Tsz To Lo, Tsun Sun Li, King Chak Wong, Wei Xue, R. Trillo","doi":"10.1609/aaai.v37i13.26907","DOIUrl":"https://doi.org/10.1609/aaai.v37i13.26907","url":null,"abstract":"The significant development of artificial neural network architectures has facilitated the increasing adoption of automated music composition models over the past few years. However, most existing systems feature algorithmic generative structures based on hard code and predefined rules, generally excluding interactive or improvised behaviors. We propose a motion based music system, MoMusic, as a AI real time music generation system. MoMusic features a partially randomized harmonic sequencing model based on a probabilistic analysis of tonal chord progressions, mathematically abstracted through musical set theory. This model is presented against a dual dimension grid that produces resulting sounds through a posture recognition mechanism. A camera captures the users' fingers' movement and trajectories, creating coherent, partially improvised harmonic progressions. MoMusic integrates several timbrical registers, from traditional classical instruments such as the piano to a new ''human voice instrument'' created using a voice conversion technique. Our research demonstrates MoMusic's interactiveness, ability to inspire musicians, and ability to generate coherent musical material with various timbrical registers. MoMusic's capabilities could be easily expanded to incorporate different forms of posture controlled timbrical transformation, rhythmic transformation, dynamic transformation, or even digital sound processing techniques.","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"28 1","pages":"16057-16062"},"PeriodicalIF":0.0,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74526568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-26DOI: 10.1609/aaai.v37i13.26912
Yubo Wang, Fengzhou Pan, Danni Liu, Jiaxiong Hu
While music is made to convey messages and emotions, auditory music is not equally accessible to everyone. Music visualization is a common approach to augment the listening experiences of the hearing users and to provide music experiences for the hearing-impaired. In this paper, we present a music visualization system that can turn the input of a piece of music into a series of facial expressions representative of the continuously changing sentiments in the music. The resulting facial expressions, recorded as action units, can later animate a static virtual avatar to be emotive synchronously with the music.
{"title":"Music-to-Facial Expressions: Emotion-Based Music Visualization for the Hearing Impaired","authors":"Yubo Wang, Fengzhou Pan, Danni Liu, Jiaxiong Hu","doi":"10.1609/aaai.v37i13.26912","DOIUrl":"https://doi.org/10.1609/aaai.v37i13.26912","url":null,"abstract":"While music is made to convey messages and emotions, auditory music is not equally accessible to everyone. Music visualization is a common approach to augment the listening experiences of the hearing users and to provide music experiences for the hearing-impaired. In this paper, we present a music visualization system that can turn the input of a piece of music into a series of facial expressions representative of the continuously changing sentiments in the music. The resulting facial expressions, recorded as action units, can later animate a static virtual avatar to be emotive synchronously with the music.","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"43 2","pages":"16096-16102"},"PeriodicalIF":0.0,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72482483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-26DOI: 10.1609/aaai.v37i11.26556
Pingsheng Liu, Zhengjie Huang, Xiechi Zhang, Linlin Wang, Gerard de Melo, Xin Lin, Liang Pang, Liang He
Endowing dialogue agents with personas is the key to delivering more human-like conversations. However, existing persona-grounded dialogue systems still lack informative details of human conversations and tend to reply with inconsistent and generic responses. One of the main underlying causes is that pre-defined persona sentences are generally short and merely superficial descriptions of personal attributes, making appropriate persona selection and understanding non-trivial. Another challenge is that it is crucial to consider the context and the conversation flow to dynamically determine when to invoke different types of persona signals. To address these problems, we propose a disentangled-attention based pre-training architecture, which incorporates persona-aware prompt learning to bridge the connection between the selected persona and response generation. Our model first exploits the conversation flow to select context-relevant personas, and subsequently enriches the superficial persona descriptions with extra personality traits through persona-aware prompting. Finally, the decoder leverages a disentangled-attention mechanism to flexibly control the reliance on personas and dialogue contexts, and incorporates A*-like keyword-based heuristic estimates for controllable generation. Extensive experiments show that our approach can outperform strong baselines and deliver more consistent and engaging responses on the PERSONA-CHAT dataset.
{"title":"A Disentangled-Attention Based Framework with Persona-Aware Prompt Learning for Dialogue Generation","authors":"Pingsheng Liu, Zhengjie Huang, Xiechi Zhang, Linlin Wang, Gerard de Melo, Xin Lin, Liang Pang, Liang He","doi":"10.1609/aaai.v37i11.26556","DOIUrl":"https://doi.org/10.1609/aaai.v37i11.26556","url":null,"abstract":"Endowing dialogue agents with personas is the key to delivering more human-like conversations. However, existing persona-grounded dialogue systems still lack informative details of human conversations and tend to reply with inconsistent and generic responses. One of the main underlying causes is that pre-defined persona sentences are generally short and merely superficial descriptions of personal attributes, making appropriate persona selection and understanding non-trivial. Another challenge is that it is crucial to consider the context and the conversation flow to dynamically determine when to invoke different types of persona signals. To address these problems, we propose a disentangled-attention based pre-training architecture, which incorporates persona-aware prompt learning to bridge the connection between the selected persona and response generation. Our model first exploits the conversation flow to select context-relevant personas, and subsequently enriches the superficial persona descriptions with extra personality traits through persona-aware prompting. Finally, the decoder leverages a disentangled-attention mechanism to flexibly control the reliance on personas and dialogue contexts, and incorporates A*-like keyword-based heuristic estimates for controllable generation. Extensive experiments show that our approach can outperform strong baselines and deliver more consistent and engaging responses on the PERSONA-CHAT dataset.","PeriodicalId":74506,"journal":{"name":"Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence","volume":"10 1","pages":"13255-13263"},"PeriodicalIF":0.0,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72581913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号