Filters








3,266 Hits in 5.2 sec

Unified Semantic Parsing with Weak Supervision

Priyanka Agrawal, Ayushi Dalmia, Parag Jain, Abhishek Bansal, Ashish Mittal, Karthik Sankaranarayanan
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
To overcome this, we propose a novel framework to build a unified multi-domain enabled semantic parser trained only with weak supervision (denotations).  ...  To solve this, we incorporate a multipolicy distillation mechanism in which we first train domain-specific semantic parsers (teachers) using weak supervision in the absence of the ground truth programs  ...  Note that: (1) Our teachers are trained with weak supervision from denotations instead of actual parses and hence are weaker compared to completely supervised semantic parses. (2) Stateof-the-art sequence  ... 
doi:10.18653/v1/p19-1473 dblp:conf/acl/AgrawalDJBMS19 fatcat:shwrsvoqfbag7kzhtrc7gvvm64

Semantic Parsing with Combinatory Categorial Grammars

Yoav Artzi, Nicholas FitzGerald, Luke S. Zettlemoyer
2013 Annual Meeting of the Association for Computational Linguistics  
Semantic parsers map natural language sentences to formal representations of their underlying meaning.  ...  Building accurate semantic parsers without prohibitive engineering costs is a longstanding, open research problem.  ...  The approach learns from data with labeled meaning representations, as well as from more easily gathered weak supervision.  ... 
dblp:conf/acl/ArtziFZ13 fatcat:jjswljcn6naifbedxcvtobyxji

Statistical learning for semantic parsing: A survey

Qile Zhu, Xiyao Ma, Xiaolin Li
2019 Big Data Mining and Analytics  
With the rise of deep learning, we will pay more attention on the deep learning based semantic parsing, especially for the application of Knowledge Base Question Answering (KBQA).  ...  One way to achieve this goal is semantic parsing.  ...  Deep Learning for Semantic Parsing In this section, we will first give an overview of deep learning, and then focus on the deep learning based semantic parsing algorithms for both supervised and weak supervised  ... 
doi:10.26599/bdma.2019.9020011 dblp:journals/bigdatama/ZhuML19 fatcat:evuhlbbl7jd67ajemd6xolxrje

Toward Code Generation: A Survey and Lessons from Semantic Parsing [article]

Celine Lee
2021 arXiv   pre-print
We then consider semantic parsing works from an evolutionary perspective, with specific analyses on neuro-symbolic methods, architecture, and supervision.  ...  We begin by reviewing natural language semantic parsing techniques and draw parallels with program synthesis efforts.  ...  Supervision in Semantic Parsing Parallel to the evolution of NL semantic parsing techniques is the evolution of supervision for semantic parsing.  ... 
arXiv:2105.03317v1 fatcat:34zbwwxrfbhuth7pk2vquq6vaq

Deep Structured Scene Parsing by Learning with Image Descriptions [article]

Liang Lin, Guangrun Wang, Rui Zhang, Ruimao Zhang, Xiaodan Liang, Wangmeng Zuo
2018 arXiv   pre-print
This paper addresses a fundamental problem of scene understanding: How to parse the scene image into a structured configuration (i.e., a semantic object hierarchy with object interaction relations) that  ...  finely accords with human perception.  ...  Weakly-supervised Model Training Compared with some other weak annotations such as labels and attributes, sentences usually provide richer semantics and structured contexts (e.g., object interactions and  ... 
arXiv:1604.02271v3 fatcat:laaiypmr7bcprixffwbgku22ja

Maximum Margin Reward Networks for Learning from Explicit and Implicit Supervision

Haoruo Peng, Ming-Wei Chang, Wen-tau Yih
2017 Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing  
On named entity recognition and semantic parsing, our model outperforms previous systems on the benchmark datasets, CoNLL-2003 and WebQuestionsSP. KB KB  ...  Neural networks have achieved state-ofthe-art performance on several structuredoutput prediction tasks, trained in a fully supervised fashion.  ...  Our system trained with MMRN is comparable to the state-of-art NTEL system. Table 3 : 3 Implicit Supervision: Semantic Parsing.  ... 
doi:10.18653/v1/d17-1252 dblp:conf/emnlp/PengCY17 fatcat:qs2ss7awszbqnc4h547vw2fghu

Weakly Supervised Graph Propagation Towards Collective Image Parsing

Si Liu, Shuicheng Yan, Tianzhu Zhang, Changsheng Xu, Jing Liu, Hanqing Lu
2012 IEEE transactions on multimedia  
In this work, we propose a weakly supervised graph propagation method to automatically assign the annotated labels at image level to those contextually derived semantic regions.  ...  Image-level labels are imposed on the graph as weak supervision information over subgraphs, each of which corresponds to all patches of one image, and the contextual information across different images  ...  are semantic consistent regions with corresponding labels.  ... 
doi:10.1109/tmm.2011.2174780 fatcat:6i3fk4vfrjho3o7s3uw6w43joa

Deep Structured Scene Parsing by Learning with Image Descriptions

Liang Lin, Guangrun Wang, Rui Zhang, Ruimao Zhang, Xiaodan Liang, Wangmeng Zuo
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
This paper addresses a fundamental problem of scene understanding: How to parse the scene image into a structured configuration (i.e., a semantic object hierarchy with object interaction relations) that  ...  finely accords with human perception.  ...  Weakly-supervised Model Training Compared with some other weak annotations such as labels and attributes, sentences usually provide richer semantics and structured contexts (e.g., object interactions and  ... 
doi:10.1109/cvpr.2016.250 dblp:conf/cvpr/LinWZZLZ16 fatcat:6gwwarmjk5fcvj5nentazu2vhq

Hierarchical Scene Parsing by Weakly Supervised Learning with Image Descriptions [article]

Ruimao Zhang, Liang Lin, Guangrun Wang, Meng Wang, Wangmeng Zuo
2018 arXiv   pre-print
In particular, SYSU-Scenes contains more than 5000 scene images with their semantic sentence descriptions, which is created by us for advancing research on scene parsing.  ...  This paper investigates a fundamental problem of scene understanding: how to parse a scene image into a structured configuration (i.e., a semantic object hierarchy with object interaction relations).  ...  On PASCAL VOC 2012, compared with our weakly supervised CNN-RsNN baseline, the improvement on IoU is 8.6% with 280 strongly annotated images (amount of "strong" : "weak" samples = 1:5), and is 16.6% with  ... 
arXiv:1709.09490v2 fatcat:nzxb246g7ranniwzawl23jnjie

Unified Multisensory Perception: Weakly-Supervised Audio-Visual Video Parsing [article]

Yapeng Tian, Dingzeyu Li, Chenliang Xu
2020 arXiv   pre-print
Experimental results show that the challenging audio-visual video parsing can be achieved even with only video-level weak labels.  ...  To facilitate exploration, we collect a Look, Listen, and Parse (LLP) dataset to investigate audio-visual video parsing in a weakly-supervised manner.  ...  with semantic labels.  ... 
arXiv:2007.10558v1 fatcat:kcexne6cpbe2tfyimttmercbka

On Symbiosis of Attribute Prediction and Semantic Segmentation [article]

Mahdi M. Kalayeh, Mubarak Shah
2019 arXiv   pre-print
Therefore, in addition to prediction, we are able to localize the attributes despite merely having access to image-level labels (weak supervision) during training.  ...  We build our attribute prediction model jointly with a deep semantic segmentation network.  ...  Therefore, it is easy to see that when few training instances are available, indeed image-level facial attribute labels can serve as an effective source of weak supervision to improve semantic face parsing  ... 
arXiv:1911.11612v1 fatcat:vz5lqkq7bfftpmcimov2hnjts4

Multi-class Semantic Video Segmentation with Exemplar-Based Object Reasoning

Buyu Liu, Xuming He, Stephen Gould
2015 2015 IEEE Winter Conference on Applications of Computer Vision  
We demonstrate the effectiveness of our method on three public datasets and show that our model can achieve superior or comparable results than the stateof-the-art with less object-level supervision.  ...  We tackle the problem of semantic segmentation of dynamic scene in video sequences.  ...  Table 4 shows the detector performance under weak supervision.  ... 
doi:10.1109/wacv.2015.140 dblp:conf/wacv/LiuHG15 fatcat:pumpr6xq3ndxrc4nx4rqmwf6pe

Relation Extraction Using TBL with Distant Supervision

Maengsik Choi, Harksoo Kim
2014 Joint International Conference of Semantic Technology  
supervision method.  ...  Supervised machine learning methods have been widely used in relation extraction that finds the relation between two named entities in a sentence.  ...  Construction of Weakly Labeled Data with Distant Supervision For distant supervision, we use DBpedia ontology as a knowledge base.  ... 
dblp:conf/jist/ChoiK14 fatcat:24ejwgshfjecfm3dmv63rlhjle

Complex Knowledge Base Question Answering: A Survey [article]

Yunshi Lan, Gaole He, Jinhao Jiang, Jing Jiang, Wayne Xin Zhao, Ji-Rong Wen
2022 arXiv   pre-print
Next, we present two mainstream categories of methods for complex KBQA, namely semantic parsing-based (SP-based) methods and information retrieval-based (IR-based) methods.  ...  In detail, we begin with introducing the complex KBQA task and relevant background.  ...  However, the insufficient training data makes it a challenge to train under weak supervision.  ... 
arXiv:2108.06688v2 fatcat:frcdrrhbsncm3kprehnz563yfq

Surveillance Video Parsing with Single Frame Supervision

Si Liu, Changhu Wang, Ruihe Qian, Han Yu, Renda Bao, Yao Sun
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
SVP (i) roughly parses the frames within the video segment, (ii) estimates the optical flow between frames and (iii) fuses the rough parsing results warped by optical flow to produce the refined parsing  ...  To parse one particular frame, the video segment preceding the frame is jointly considered.  ...  [17] address the problem of automatically parsing the fashion images with weak supervision from the user-generated color-category tags.  ... 
doi:10.1109/cvpr.2017.114 dblp:conf/cvpr/LiuWQYBS17 fatcat:3yqf4z3zifgz5iptwf5l4k57ka
« Previous Showing results 1 — 15 out of 3,266 results