Pub Date : 1900-01-01DOI: 10.18653/v1/2021.splurobonlp-1.3
T. Dong, Alberto Testoni, Luciana Benotti, R. Bernardi
In this paper, we define and evaluate a methodology for extracting history-dependent spatial questions from visual dialogues. We say that a question is history-dependent if it requires (parts of) its dialogue history to be interpreted. We argue that some kinds of visual questions define a context upon which a follow-up spatial question relies. We call the question that restricts the context: trigger, and we call the spatial question that requires the trigger question to be answered: zoomer. We automatically extract different trigger and zoomer pairs based on the visual property that the questions rely on (e.g. color, number). We manually annotate the automatically extracted trigger and zoomer pairs to verify which zoomers require their trigger. We implement a simple baseline architecture based on a SOTA multimodal encoder. Our results reveal that there is much room for improvement for answering history-dependent questions.
{"title":"Visually Grounded Follow-up Questions: a Dataset of Spatial Questions Which Require Dialogue History","authors":"T. Dong, Alberto Testoni, Luciana Benotti, R. Bernardi","doi":"10.18653/v1/2021.splurobonlp-1.3","DOIUrl":"https://doi.org/10.18653/v1/2021.splurobonlp-1.3","url":null,"abstract":"In this paper, we define and evaluate a methodology for extracting history-dependent spatial questions from visual dialogues. We say that a question is history-dependent if it requires (parts of) its dialogue history to be interpreted. We argue that some kinds of visual questions define a context upon which a follow-up spatial question relies. We call the question that restricts the context: trigger, and we call the spatial question that requires the trigger question to be answered: zoomer. We automatically extract different trigger and zoomer pairs based on the visual property that the questions rely on (e.g. color, number). We manually annotate the automatically extracted trigger and zoomer pairs to verify which zoomers require their trigger. We implement a simple baseline architecture based on a SOTA multimodal encoder. Our results reveal that there is much room for improvement for answering history-dependent questions.","PeriodicalId":442464,"journal":{"name":"Proceedings of Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics","volume":"126 26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128026245","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 : 1900-01-01DOI: 10.18653/v1/2021.splurobonlp-1.1
Mattias Appelgren, A. Lascarides
This paper describes a method for learning from a teacher’s potentially unreliable corrective feedback in an interactive task learning setting. The graphical model uses discourse coherence to jointly learn symbol grounding, domain concepts and valid plans. Our experiments show that the agent learns its domain-level task in spite of the teacher’s mistakes.
{"title":"Symbol Grounding and Task Learning from Imperfect Corrections","authors":"Mattias Appelgren, A. Lascarides","doi":"10.18653/v1/2021.splurobonlp-1.1","DOIUrl":"https://doi.org/10.18653/v1/2021.splurobonlp-1.1","url":null,"abstract":"This paper describes a method for learning from a teacher’s potentially unreliable corrective feedback in an interactive task learning setting. The graphical model uses discourse coherence to jointly learn symbol grounding, domain concepts and valid plans. Our experiments show that the agent learns its domain-level task in spite of the teacher’s mistakes.","PeriodicalId":442464,"journal":{"name":"Proceedings of Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131765177","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 : 1900-01-01DOI: 10.18653/v1/2021.splurobonlp-1.8
Haein Jeon, Yewon Kim, Bo-Yeong Kang
With the development of robotics, the use of robots in daily life is increasing, which has led to the need for anyone to easily train robots to improve robot use. Interactive reinforcement learning(IARL) is a method for robot training based on human–robot interaction; prior studies on IARL provide only limited types of feedback or require appropriately designed shaping rewards, which is known to be difficult and time-consuming. Therefore, in this study, we propose interactive deep reinforcement learning models based on voice feedback. In the proposed system, a robot learns the task of cooperative table balancing through deep Q-network using voice feedback provided by humans in real-time, with automatic speech recognition(ASR) and sentiment analysis to understand human voice feedback. As a result, an optimal policy convergence rate of up to 96% was realized, and performance was improved in all voice feedback-based models
{"title":"Interactive Reinforcement Learning for Table Balancing Robot","authors":"Haein Jeon, Yewon Kim, Bo-Yeong Kang","doi":"10.18653/v1/2021.splurobonlp-1.8","DOIUrl":"https://doi.org/10.18653/v1/2021.splurobonlp-1.8","url":null,"abstract":"With the development of robotics, the use of robots in daily life is increasing, which has led to the need for anyone to easily train robots to improve robot use. Interactive reinforcement learning(IARL) is a method for robot training based on human–robot interaction; prior studies on IARL provide only limited types of feedback or require appropriately designed shaping rewards, which is known to be difficult and time-consuming. Therefore, in this study, we propose interactive deep reinforcement learning models based on voice feedback. In the proposed system, a robot learns the task of cooperative table balancing through deep Q-network using voice feedback provided by humans in real-time, with automatic speech recognition(ASR) and sentiment analysis to understand human voice feedback. As a result, an optimal policy convergence rate of up to 96% was realized, and performance was improved in all voice feedback-based models","PeriodicalId":442464,"journal":{"name":"Proceedings of Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127144678","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 : 1900-01-01DOI: 10.18653/v1/2021.splurobonlp-1.9
Sayali Kulkarni, Shailee Jain, Mohammad Javad Hosseini, Jason Baldridge, Eugene Ie, Li Zhang
We present a multi-level geocoding model (MLG) that learns to associate texts to geographic coordinates. The Earth’s surface is represented using space-filling curves that decompose the sphere into a hierarchical grid. MLG balances classification granularity and accuracy by combining losses across multiple levels and jointly predicting cells at different levels simultaneously. It obtains large gains without any gazetteer metadata, demonstrating that it can effectively learn the connection between text spans and coordinates—and thus makes it a gazetteer-free geocoder. Furthermore, MLG obtains state-of-the-art results for toponym resolution on three English datasets without any dataset-specific tuning.
{"title":"Multi-Level Gazetteer-Free Geocoding","authors":"Sayali Kulkarni, Shailee Jain, Mohammad Javad Hosseini, Jason Baldridge, Eugene Ie, Li Zhang","doi":"10.18653/v1/2021.splurobonlp-1.9","DOIUrl":"https://doi.org/10.18653/v1/2021.splurobonlp-1.9","url":null,"abstract":"We present a multi-level geocoding model (MLG) that learns to associate texts to geographic coordinates. The Earth’s surface is represented using space-filling curves that decompose the sphere into a hierarchical grid. MLG balances classification granularity and accuracy by combining losses across multiple levels and jointly predicting cells at different levels simultaneously. It obtains large gains without any gazetteer metadata, demonstrating that it can effectively learn the connection between text spans and coordinates—and thus makes it a gazetteer-free geocoder. Furthermore, MLG obtains state-of-the-art results for toponym resolution on three English datasets without any dataset-specific tuning.","PeriodicalId":442464,"journal":{"name":"Proceedings of Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121984488","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：