Haw-Shiuan Chang | Publications

My Semantic Scholar and Google Scholar pages

2022

ACL

Softmax Bottleneck Makes Language Models Unable to Represent Multi-mode Word Distributions

Haw-Shiuan Chang, and Andrew McCallum

In Annual Meeting of the Association for Computational Linguistics (ACL) 2022

Paper Talk Poster Slides Code Bib

@inproceedings{chang2022softmax,
  title = {Softmax Bottleneck Makes Language Models Unable to Represent Multi-mode Word Distributions},
  abbr = {ACL},
  selected = {true},
  bibtex_show = {true},
  author = {Chang, Haw-Shiuan and McCallum, Andrew},
  booktitle = {Annual Meeting of the Association for Computational Linguistics (ACL)},
  year = {2022},
  paper = {https://aclanthology.org/2022.acl-long.554.pdf},
  code = {https://aclanthology.org/attachments/2022.acl-long.554.software.zip},
  poster = {ACL2022_poster.pdf},
  slides = {ACL2022_slides.pdf},
  video = {https://screencast-o-matic.com/watch/c3fresVYo4M}
}

2021

EACL Oral

Changing the Mind of Transformers for Topically-Controllable Language Generation

Haw-Shiuan Chang, Jiaming Yuan, Mohit Iyyer, and Andrew McCallum

In Conference of the European Chapter of the Association for Computational Linguistics (EACL) (Oral) 2021

Paper Talk Poster Slides Code Bib

@inproceedings{chang2021changing,
  title = {Changing the Mind of Transformers for Topically-Controllable Language Generation},
  abbr = {EACL Oral},
  selected = {true},
  bibtex_show = {true},
  author = {Chang, Haw-Shiuan and Yuan, Jiaming and Iyyer, Mohit and McCallum, Andrew},
  booktitle = {Conference of the European Chapter of the Association for Computational Linguistics (EACL) (Oral)},
  year = {2021},
  paper = {https://arxiv.org/abs/2103.15335},
  code = {https://github.com/iesl/interactive_LM},
  video = {https://slideslive.com/38954487/changing-the-mind-of-transformers-for-topicallycontrollable-language-generation},
  poster = {proj1-nlp-interactive_LM-EACL_interactive_LM-poster.pdf},
  slides = {proj1-nlp-interactive_LM-EACL_interactive_LM-slides.key}
}

EACL Oral

Multi-facet Universal Schema

Rohan Paul*, Haw-Shiuan Chang*, and Andrew McCallum

In Conference of the European Chapter of the Association for Computational Linguistics (EACL) (Oral) 2021

Paper Talk Poster Slides Code Bib

@inproceedings{chang2021multi-facet,
  title = {Multi-facet Universal Schema},
  abbr = {EACL Oral},
  bibtex_show = {true},
  author = {Paul*, Rohan and Chang*, Haw-Shiuan and McCallum, Andrew},
  booktitle = {Conference of the European Chapter of the Association for Computational Linguistics (EACL) (Oral)},
  year = {2021},
  paper = {https://arxiv.org/abs/2103.15339},
  code = {https://github.com/rohanpaul11/multifacet-re},
  video = {https://slideslive.com/38954382/multifacet-universal-schema},
  poster = {proj2-nlp-Multi-facet RE-EACL_multi-facet_RE_small-poster.pdf},
  slides = {proj2-nlp-Multi-facet RE-EACL_multi-facet_RE_small-slides.key}
}

AAAI

Extending Multi-Sense Word Embedding to Phrases and Sentences for Unsupervised Semantic Applications

Haw-Shiuan Chang, Amol Agrawal, and Andrew McCallum

In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI) 2021

Paper Poster Slides Bib

@inproceedings{chang2021extending,
  title = {Extending Multi-Sense Word Embedding to Phrases and Sentences for Unsupervised Semantic Applications},
  abbr = {AAAI},
  bibtex_show = {true},
  author = {Chang, Haw-Shiuan and Agrawal, Amol and McCallum, Andrew},
  booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2021},
  paper = {https://arxiv.org/abs/2103.15330},
  poster = {proj3-nlp-Multi-facet Embeddings for Sentence Representation-poster.pdf},
  slides = {https://docs.google.com/presentation/d/1k-OBWdBYsGmXUuvNc1_J_JrEppB_Aqh82bJPCgjFL-s/edit?usp=sharing}
}

EMNLP Short

Open Aspect Target Sentiment Classification with Natural Language Prompts

Ronald Seoh*, Ian Birle*, Mrinal Tak*, Haw-Shiuan Chang*, Brian Pinette, and Alfred Hough

In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP) 2021

Paper Bib

@inproceedings{Seoh_EMNLP_21,
  title = {Open Aspect Target Sentiment Classification with Natural Language Prompts},
  abbr = {EMNLP Short},
  bibtex_show = {true},
  author = {Seoh*, Ronald and Birle*, Ian and Tak*, Mrinal and Chang*, Haw-Shiuan and Pinette, Brian and Hough, Alfred},
  booktitle = {Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  paper = {https://arxiv.org/abs/2109.03685},
  year = {2021}
}

2020

Using Error Decay Prediction to Overcome Practical Issues of Deep Active Learning for Named Entity Recognition

Haw-Shiuan Chang, Shankar Vembu, Sunil Mohan, Rheeya Uppaal, and Andrew McCallum

Machine Learning 2020

Paper Talk Slides Bib

@article{chang2019ovecoming,
  abbr = {ML},
  bibtex_show = {true},
  selected = {true},
  author = {Chang, Haw-Shiuan and Vembu, Shankar and Mohan, Sunil and Uppaal, Rheeya and McCallum, Andrew},
  journal = {Machine Learning},
  paper = {http://arxiv.org/abs/1911.07335},
  title = {Using Error Decay Prediction to Overcome Practical Issues of Deep Active Learning for Named Entity Recognition},
  year = {2020},
  doi = {10.1007/s10994-020-05897-1},
  publisher = {Springer},
  video = {https://slideslive.com/38933012/using-error-decay-prediction-to-overcome-practical-issues-of-deep-active-learning-for-named-entity-recognition},
  slides = {https://docs.google.com/presentation/d/1h3bI1dS8-vS5ItcfGOHN3J7Wi7dnWWBf0gA4-yBXrk4/edit?usp=sharing}
}

KDD

AutoKnow: Self-Driving Knowledge Collection for Products of Thousands of Types

Xin Luna Dong, Xiang He, Andrey Kan, Xian Li, Yan Liang, Jun Ma, Yifan Ethan Xu, Chenwei Zhang, Tong Zhao, Gabriel Blanco Saldana, Saurabh Deshpande, Alexandre Michetti Manduca, Jay Ren, Surender Pal Singh, Fan Xiao, Haw-Shiuan Chang, Giannis Karamanolakis, Yuning Mao, Yaqing Wang, Christos Faloutsos, Andrew McCallum, and Jiawei Han

In KDD ’20: The 26th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Virtual Event, CA, USA, August 23-27, 2020 2020

Paper Bib

@inproceedings{DBLP:conf/kdd/DongHKLLMXZZSDM20,
  author = {Dong, Xin Luna and He, Xiang and Kan, Andrey and Li, Xian and Liang, Yan and Ma, Jun and Xu, Yifan Ethan and Zhang, Chenwei and Zhao, Tong and Saldana, Gabriel Blanco and Deshpande, Saurabh and Manduca, Alexandre Michetti and Ren, Jay and Singh, Surender Pal and Xiao, Fan and Chang, Haw{-}Shiuan and Karamanolakis, Giannis and Mao, Yuning and Wang, Yaqing and Faloutsos, Christos and McCallum, Andrew and Han, Jiawei},
  abbr = {KDD},
  bibtex_show = {true},
  title = {AutoKnow: Self-Driving Knowledge Collection for Products of Thousands
                 of Types},
  booktitle = {{KDD} '20: The 26th {ACM} {SIGKDD} Conference on Knowledge Discovery
                 and Data Mining, Virtual Event, CA, USA, August 23-27, 2020},
  pages = {2724--2734},
  publisher = {{ACM}},
  year = {2020},
  url = {https://doi.org/10.1145/3394486.3403323},
  doi = {10.1145/3394486.3403323},
  paper = {https://arxiv.org/abs/2006.13473}
}

JCIM

Inorganic Materials Synthesis Planning with Literature-Trained Neural Networks

Edward Kim, Zach Jensen, Alexander Grootel, Kevin Huang, Matthew Staib, Sheshera Mysore, Haw-Shiuan Chang, Emma Strubell, Andrew McCallum, Stefanie Jegelka, and Elsa Olivetti

J. Chem. Inf. Model. 2020

Paper Bib

@article{DBLP:journals/jcisd/KimJGHSMCSMJO20,
  author = {Kim, Edward and Jensen, Zach and van Grootel, Alexander and Huang, Kevin and Staib, Matthew and Mysore, Sheshera and Chang, Haw{-}Shiuan and Strubell, Emma and McCallum, Andrew and Jegelka, Stefanie and Olivetti, Elsa},
  abbr = {JCIM},
  bibtex_show = {true},
  title = {Inorganic Materials Synthesis Planning with Literature-Trained Neural
                 Networks},
  journal = {J. Chem. Inf. Model.},
  volume = {60},
  number = {3},
  pages = {1194--1201},
  year = {2020},
  url = {https://doi.org/10.1021/acs.jcim.9b00995},
  doi = {10.1021/acs.jcim.9b00995},
  paper = {https://arxiv.org/abs/1901.00032}
}

2019

LAW

The Materials Science Procedural Text Corpus: Annotating Materials Synthesis Procedures with Shallow Semantic Structures

Sheshera Mysore, Zach Jensen, Edward Kim, Kevin Huang, Haw-Shiuan Chang, Emma Strubell, Jeffrey Flanigan, Andrew McCallum, and Elsa Olivetti

In Proceedings of the 13th Linguistic Annotation Workshop at ACL 2019

Paper Bib

@inproceedings{mysore2019msannlaw,
  title = {The Materials Science Procedural Text Corpus: Annotating Materials Synthesis Procedures with Shallow Semantic Structures},
  abbr = {LAW},
  bibtex_show = {true},
  author = {Mysore, Sheshera and Jensen, Zach and Kim, Edward and Huang, Kevin and Chang, Haw-Shiuan and Strubell, Emma and Flanigan, Jeffrey and McCallum, Andrew and Olivetti, Elsa},
  booktitle = {Proceedings of the 13th Linguistic Annotation Workshop at ACL},
  year = {2019},
  paper = {https://sigann.github.io/LAW-XIII-2019/pdf/W19-4007.pdf}
}

2018

TMM

Active Learning for Crowdsourced QoE Modeling

Haw-Shiuan Chang, Chih-Fan Hsu, Tobias Hoßfeld, and Kuan-Ta Chen

IEEE Trans. Multim. 2018

Paper Bib

@article{DBLP:journals/tmm/ChangHHC18,
  abbr = {TMM},
  bibtex_show = {true},
  author = {Chang, Haw{-}Shiuan and Hsu, Chih{-}Fan and Ho{\ss}feld, Tobias and Chen, Kuan{-}Ta},
  title = {Active Learning for Crowdsourced QoE Modeling},
  journal = {{IEEE} Trans. Multim.},
  volume = {20},
  number = {12},
  pages = {3337--3352},
  year = {2018},
  paper = {proj9-cv-Active Sampling for estimating QoE model-paper.pdf}
}

NAACL

Distributional Inclusion Vector Embedding for Unsupervised Hypernymy Detection

Haw-Shiuan Chang, ZiYun Wang, Luke Vilnis, and Andrew McCallum

In Human Language Technology Conference of the North American Chapter of the Association of Computational Linguistics (HLT/NAACL) 2018

Paper Demo Poster Code Bib

@inproceedings{chang2017unsupervised,
  title = {Distributional Inclusion Vector Embedding for Unsupervised Hypernymy Detection},
  abbr = {NAACL},
  bibtex_show = {true},
  selected = {true},
  author = {Chang, Haw-Shiuan and Wang, ZiYun and Vilnis, Luke and McCallum, Andrew},
  booktitle = {Human Language Technology Conference of the North American Chapter of the Association of Computational Linguistics (HLT/NAACL)},
  paper = {http://arxiv.org/abs/1710.00880},
  year = {2018},
  code = {https://github.com/iesl/Distributional-Inclusion-Vector-Embedding},
  poster = {proj5-nlp-DIVE-poster.pdf},
  demo = {https://bl.ocks.org/chsu5358/raw/f08d4755b0f04e113c139a72a977df5c/}
}

TextGraphs

Efficient Graph-based Word Sense Induction by Distributional Inclusion Vector Embeddings

Haw-Shiuan Chang, Amol Agrawal, Ananya Ganesh, Anirudha Desai, Vinayak Mathur, Alfred Hough, and Andrew McCallum

In TextGraphs-12: the Workshop on Graph-based Methods for Natural Language Processing (NAACL WS) 2018

Paper Slides Bib

@inproceedings{conf/TextGraph18/Chang18,
  abbr = {TextGraphs},
  bibtex_show = {true},
  author = {Chang, Haw-Shiuan and Agrawal, Amol and Ganesh, Ananya and Desai, Anirudha and Mathur, Vinayak and Hough, Alfred and McCallum, Andrew},
  title = {Efficient Graph-based Word Sense Induction by Distributional Inclusion Vector Embeddings},
  booktitle = {TextGraphs-12: the Workshop on Graph-based Methods for Natural Language Processing (NAACL WS)},
  year = {2018},
  slides = {proj5-nlp-DIVE-slides.pdf},
  paper = {https://arxiv.org/abs/1804.03257}
}

2017

NIPS

Active Bias: Training a More Accurate Neural Network by Emphasizing High Variance Samples

Haw-Shiuan Chang, Erik G. Learned-Miller, and Andrew McCallum

In Advances in Neural Information Processing Systems (NIPS) 2017

Paper Poster Bib

@inproceedings{DBLP:conf/nips/ChangLM17,
  abbr = {NIPS},
  bibtex_show = {true},
  selected = {true},
  author = {Chang, Haw{-}Shiuan and Learned{-}Miller, Erik G. and McCallum, Andrew},
  title = {Active Bias: Training a More Accurate Neural Network by Emphasizing
                 High Variance Samples},
  booktitle = {Advances in Neural Information Processing Systems (NIPS)},
  year = {2017},
  poster = {http://people.umass.edu/hawshiuancha/NIPS_poster_active_bias.pdf},
  paper = {http://arxiv.org/abs/1704.07433}
}

NIPS WS

Automatically Extracting Action Graphs from Materials Science Synthesis Procedures

Sheshera Mysore, Edward Kim, Emma Strubell, Ao Liu, Haw-Shiuan Chang, Srikrishna Kompella, Kevin Huang, Andrew McCallum, and Elsa Olivetti

In Workshop on Machine Learning for Molecules and Materials at NIPS 2017

Paper Bib

@inproceedings{mysore2017automatically,
  title = {Automatically Extracting Action Graphs from Materials Science Synthesis Procedures},
  abbr = {NIPS WS},
  bibtex_show = {true},
  author = {Mysore, Sheshera and Kim, Edward and Strubell, Emma and Liu, Ao and Chang, Haw-Shiuan and Kompella, Srikrishna and Huang, Kevin and McCallum, Andrew and Olivetti, Elsa},
  booktitle = {Workshop on Machine Learning for Molecules and Materials at NIPS},
  year = {2017},
  paper = {https://arxiv.org/abs/1711.06872}
}

2016

TAC/KBP

Extracting Multilingual Relations under Limited Resources: TAC 2016 Cold-Start KB construction and Slot-Filling using Compositional Universal Schema

Haw-Shiuan Chang, Abdurrahman Munir, Ao Liu, Johnny Tian-Zheng Wei, Aaron Traylor, Ajay Nagesh, Nicholas Monath, Patrick Verga, Emma Strubell, and Andrew McCallum

In Text Analysis Conference, Knowledge Base Population (TAC/KBP) 2016

Paper Bib

@inproceedings{DBLP:conf/tac/Chang16,
  abbr = {TAC/KBP},
  bibtex_show = {true},
  author = {Chang, Haw-Shiuan and Munir, Abdurrahman and Liu, Ao and Wei, Johnny Tian-Zheng and Traylor, Aaron and Nagesh, Ajay and Monath, Nicholas and Verga, Patrick and Strubell, Emma and McCallum, Andrew},
  title = {Extracting Multilingual Relations under Limited Resources: TAC 2016 Cold-Start KB construction and Slot-Filling using Compositional Universal Schema},
  booktitle = {Text Analysis Conference, Knowledge Base Population (TAC/KBP)},
  year = {2016},
  paper = {proj6-nlp-UMASS TAC 2016-paper.pdf}
}

2015

EDM Short

Modeling Exercise Relationships in E-Learning: A Unified Approach

Haw-Shiuan Chang, Hwai-Jung Hsu, and Kuan-Ta Chen

In International Conference on Educational Data Mining (EDM) 2015

Paper Demo Dataset Slides Bib

@inproceedings{DBLP:conf/edm/ChangHC15,
  abbr = {EDM Short},
  selected = {true},
  bibtex_show = {true},
  author = {Chang, Haw{-}Shiuan and Hsu, Hwai{-}Jung and Chen, Kuan{-}Ta},
  title = {Modeling Exercise Relationships in E-Learning: A Unified Approach},
  booktitle = {International Conference on Educational Data Mining (EDM)},
  year = {2015},
  paper = {proj8-education-student modeling-paper.pdf},
  demo = {https://drive.google.com/file/d/1V5nEwmVkPclj_ZR-bKt9xzW7ApT6ZJo-/view?usp=sharing},
  slides = {proj8-education-student%20modeling-presentation.pptx},
  dataset = {https://pslcdatashop.web.cmu.edu/DatasetInfo?datasetId=1198}
}

CVIU

Optimizing the Decomposition for Multiple Foreground Cosegmentation

Haw-Shiuan Chang, and Yu-Chiang Frank Wang

Computer Vision and Image Understanding (CVIU) 2015

Paper Code Bib

@article{CHANG201518,
  abbr = {CVIU},
  bibtex_show = {true},
  title = {Optimizing the Decomposition for Multiple Foreground Cosegmentation},
  journal = {Computer Vision and Image Understanding (CVIU)},
  volume = {141},
  pages = {18-27},
  year = {2015},
  issn = {1077-3142},
  doi = {https://doi.org/10.1016/j.cviu.2015.06.004},
  url = {https://www.sciencedirect.com/science/article/pii/S1077314215001320},
  author = {Chang, Haw-Shiuan and Wang, Yu-Chiang Frank},
  code = {http://mml.citi.sinica.edu.tw/papers/MFC_code_CVIU_2015},
  paper = {proj11-cv-Decomposition of Multiple Foreground Co-segmentation-paper.pdf}
}

2014

ACCV

Simple-to-Complex Discriminative Clustering for Hierarchical Image Segmentation

Haw-Shiuan Chang, and Yu-Chiang Frank Wang

In Asian Conference on Computer Vision (ACCV) 2014

Paper Code Bib

@inproceedings{DBLP:conf/accv/ChangW14,
  abbr = {ACCV},
  bibtex_show = {true},
  author = {Chang, Haw{-}Shiuan and Wang, Yu{-}Chiang Frank},
  editor = {Cremers, Daniel and Reid, Ian D. and Saito, Hideo and Yang, Ming{-}Hsuan},
  title = {Simple-to-Complex Discriminative Clustering for Hierarchical Image
                 Segmentation},
  booktitle = {Asian Conference on Computer Vision (ACCV)},
  series = {Lecture Notes in Computer Science},
  volume = {9005},
  pages = {391--407},
  publisher = {Springer},
  year = {2014},
  url = {https://doi.org/10.1007/978-3-319-16811-1\_26},
  doi = {10.1007/978-3-319-16811-1\_26},
  code = {http://mml.citi.sinica.edu.tw/papers/HDC_code_ACCV_2014/},
  paper = {proj10-cv-Hierarchical Image Segmentation without Training-paper.pdf}
}

2013

ICIP

Superpixel-based Large Displacement Optical Flow

Haw-Shiuan Chang, and Yu-Chiang Frank Wang

In IEEE International Conference on Image Processing, ICIP 2013

Paper Code Bib

@inproceedings{DBLP:conf/icip/ChangW13,
  abbr = {ICIP},
  bibtex_show = {true},
  author = {Chang, Haw{-}Shiuan and Wang, Yu{-}Chiang Frank},
  title = {Superpixel-based Large Displacement Optical Flow},
  booktitle = {{IEEE} International Conference on Image Processing, {ICIP}},
  pages = {3835--3839},
  publisher = {{IEEE}},
  year = {2013},
  url = {https://doi.org/10.1109/ICIP.2013.6738790},
  doi = {10.1109/ICIP.2013.6738790},
  code = {http://mml.citi.sinica.edu.tw/papers/SPLDOF_code.php},
  paper = {proj12-cv-Superpixel-Based Large Displacement Optical Flow-paper.pdf}
}

TIP

Exploring Visual and Motion Saliency for Automatic Video Object Extraction

Wei-Te Li, Haw-Shiuan Chang, Kuo-Chin Lien, Hui-Tang Chang, and Yu-Chiang Frank Wang

IEEE Transactions on Image Processing 2013

Paper Bib

@article{6482623,
  abbr = {TIP},
  bibtex_show = {true},
  author = {Li, Wei-Te and Chang, Haw-Shiuan and Lien, Kuo-Chin and Chang, Hui-Tang and Wang, Yu-Chiang Frank},
  journal = {IEEE Transactions on Image Processing},
  title = {Exploring Visual and Motion Saliency for Automatic Video Object Extraction},
  year = {2013},
  volume = {22},
  number = {7},
  pages = {2600-2610},
  doi = {10.1109/TIP.2013.2253483},
  paper = {pub-cv-tip2013-Exploring_Visual_and_Motion_Saliency_for_Automatic_Video_Object_Extraction.pdf}
}