2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics.

**Whova App** 
Stay in touch with your fellow conference attendees via the [Whova App](https://whova.com/portal/webapp/nacon_202106/)

**Conference Structure**
https://2021.naacl.org/blog/conference-structure/

**Walkthrough video of how to NAACL 2021** 

Please take a moment to view this video explaining how to navigate the platform, attend sessions network with other attendees. 


<figure class="video_container">
  <iframe src="https://screencast-o-matic.com/watch/crhwbGVh3vx?v=6&ff=1&title=0&controls=1" width=640  height=350  frameborder="0" allowfullscreen="true"> </iframe>
</figure>

<br>
<br>

NAACL 2021

United States

 The live session will happen in Zoom. Please click on the Join Zoom Room button to join the session.<br><br> [![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/a30e692a18780a229ad899bd366cbe7e.png)](https://zoom.us/j/93826225229?pwd=Q0hSYkFZKzlrRkZmcFVXSDJQY0hXUT09)<br><br>

<i>**Session Chair: ** Ves Stoyanov </i><br><br>

4C-Oral: Semantics: Sentence-level Semantics and Textual Inference

**Read the paper on the folowing link:**

https://www.aclweb.org/anthology/2021.naacl-main.108/

**Abstract:**

Pre-trained language models have achieved huge success on a wide range of NLP tasks. However, contextual representations from pre-trained models contain entangled semantic and syntactic information, and therefore cannot be directly used to derive useful semantic sentence embeddings for some tasks. Paraphrase pairs offer an effective way of learning the distinction between semantics and syntax, as they naturally share semantics and often vary in syntax. In this work, we present ParaBART, a semantic sentence embedding model that learns to disentangle semantics and syntax in sentence embeddings obtained by pre-trained language models. ParaBART is trained to perform syntax-guided paraphrasing, based on a source sentence that shares semantics with the target paraphrase, and a parse tree that specifies the target syntax. In this way, ParaBART learns disentangled semantic and syntactic representations from their respective inputs with separate encoders. Experiments in English show that ParaBART outperforms state-of-the-art sentence embedding models on unsupervised semantic similarity tasks. Additionally, we show that our approach can effectively remove syntactic information from semantic sentence embeddings, leading to better robustness against syntactic variation on downstream semantic tasks.


Disentangling Semantics and Syntax in Sentence Embeddings with Pre-trained Language Models

WACV is the premier international computer vision event comprising the main conference and several co-located workshops and tutorials. With its high quality and low cost, it provides an exceptional value for students, academics and industry researchers.




IEEE WACV 2021


**Welcome video**

<iframe width=1280 height=720 frameborder="0" scrolling="no" src="https://screencast-o-matic.com/player/cre62HVc33j?ff=1&title=0&controls=1" allowfullscreen="true"></iframe>

University of Colorado WASH Virtual Symposium 2021

The Emerging Leaders Conference encourages Missouri youth to become engaged in the legislative process. Divided into two parts — civic engagement and college access — this conference gives students an opportunity to build on their leadership skills.

The conference focuses on a range of activities:

* exploring and debating community issues,
* diverse political perspectives and
* the legislative decision-making process.
* Additionally, participating students access the college-planning process and experience postsecondary socialization and events on the University of Missouri campus in Columbia.

The Emerging Leaders program is open to all rising 10th through 12th grade Missouri high school students. At the end of the experience, legislators select 3-4 conference participants for a 2-day internship the following spring.

2021 Emerging Leaders Conference

**About**
<br>The PErvasive Technologies Related to Assistive Environments (PETRA) conference is a highly interdisciplinary conference that focuses on computational and engineering approaches to improve the quality of life and enhance human performance in a wide range of settings, in the workplace, at home, in public spaces, urban environments, and other. Outcomes of this conference have a broad impact in application areas that include, manufacturing, transportation, healthcare, energy systems, security and safety, robotics, biomedicine, environment and conservation, and many others. PETRA brings together very different types of technologies to also address important social and healthcare issues for sensitive populations, such as the elderly, persons suffering from chronic conditions such as Alzheimer’s, Parkinson’s, Cerebral Palsy, and other disabilities or traumas.

<br>PETRA (which means “stone” in Greek) addresses important human living needs. From a technological perspective, PETRA motivates and inspires new scientific thrusts, innovations and discoveries to improve living environments, a particularly important issue, in an increasingly crowded and complex world where the need for inclusiveness and connectivity with the rest of the world is key. For the elderly in particular, PETRA addresses the fact that, as people grow older; they will increasingly rely on technology to be able to prolong staying in their homes and enjoy an independent life.

<br>The PETRA 2021 proceedings will be published in the ACM Digital Library as part of the ACM ICPS program. The ISBN number assigned to PETRA '21 is 978-1-4503-8792-7
<br>
<br>

PETRA2021

Welcome to the first IPSO Virtual Conference. The mission of the virtual conference is to unify the global community of professionals engaged in research and clinical care for children with cerebrovascular diseases, and chart a path forward for multinational,
multidisciplinary efforts.

![](https://s3.amazonaws.com/pf-upload-01/u-59356/0/2021-06-23/za13j2t/TEST%20PROGRAM%20101024_1.jpg)

**[Download Full Agenda](https://s3.amazonaws.com/pf-upload-01/u-59356/0/2021-06-23/za13j2t/TEST%20PROGRAM%20101024_1.jpg)**
  
    
  


  



IPSO2021

[![](https://s3.amazonaws.com/pf-upload-01/u-59356/0/2021-06-23/1723jo2/button_conference-website.png)]
(https://www.ieee-sum.org)

[![](https://s3.amazonaws.com/pf-upload-01/u-59356/0/2021-06-23/0r13jpa/button_full-conference-program.png)]
(https://sum2021.exordo.com/programme/at-a-glance)

![](https://s3.amazonaws.com/pf-upload-01/u-59356/0/2021-06-23/tj23j7g/button_program-at-a-glance.png)

![](https://s3.amazonaws.com/pf-upload-01/u-59356/0/2021-06-23/t113jje/button_online-proceedings%201.png)

**General Chairs**
![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/54f9400de6f88b50207e174f03487021.jpg)
Cristian Antonelli ![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/ff02e31cbf745b1a3d02d288f1b04651.jpg)
Brodsky

**Other Resources**
![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/9ea5ce5095179c93858437c964e06f76.png)![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/07430a0da9949772ea7d6a89ac3b36d4.png)![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/2e930de9ba6c9f5aec2bb6c116b5ea69.png)


2021 IEEE Summer Topicals Meeting Series

The recent years have seen a revolution in the science and practice of crystal growth. Novel mechanisms of nucleation and growth have decoded mysteries that have puzzled the community for ages. Elaborate processes and advanced techniques have birthed materials with unprecedented properties and exciting applications. Rationales developed within the crystal growth realm have illuminated longstanding conundrums in geology, biology, and medicine. 

These jointly-held conferences are the premier forums to discuss the huge modern advances in all aspects of crystal growth, both in the bulk and as epitaxial films. We will continue the tradition of numerous ACCGE and OMVPE meetings and bring together the topmost experts in the field from industry, academia, and government laboratories. Joining the two conferences affords an exceptional opportunity to present and discuss original results and provides a venue for newcomers to embrace the hot, new advances in science and technology. Our format is designed to maximize interaction among leading specialists and the wider crystal growth community, with oral and poster sessions as well as plenary and invited speakers to provide a broad picture of the up-to-date developments in the field.

Please register for the ACCGE-22/OMVPE-20 to be able to access the content.

ACCGE-22/OMVPE-20

ComSciConCAN is Canada’s first national science communication workshop for graduate students. This unique professional development program aims to help the next generation of leaders in STEM fields develop the skills needed to communicate their research and ideas to their peers, experts in other fields, policy makers, and the general public.

ComSciConCAN looks to bring the longstanding and successful ComSciCon framework from the United States to Canada. Since its establishment in 2013 by graduate students at Harvard University and MIT, the training opportunities offered by ComSciCon have benefited over 500 graduate students across the United States and Canada. What was once a single annual national meeting has expanded to numerous local ComSciCon workshops across the United States. In collaboration with other alumni and invited experts, former ComSciCon attendees have published over 80 articles written at these workshops, initiated new science communication endeavours, and built lasting professional networks.

<br>
<br>

![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/3dcb3046cf83a8224bc6f09b74346187.png)


ComSciConCAN

CAN-ACN 2021

It is our great pleasure to invite you to attend the 26th International Workshop on Rare Earth and Future Permanent Magnets and Their Applications (REPM2021) which will be held virtually on June 7-10, 2021. Please note, this is an updated schedule due to COVID-19 travel restrictions which forced us to postpone the conference from summer of 2020. The REPM workshop takes place every two years and is the premier conference on permanent magnets covering a wide range of topics including:

* Rare earth magnets
* Rare earth-free magnets
* Raw materials, mining and recycling
* Advanced synthesis, processing and characterization techniques
* Theoretical modeling of coercivity
* Fundamental properties of advanced magnetic materials
* Thin film magnets
* Applications of permanent magnets

***Workshop***
The workshop will be chaired by Professor George C. Hadjipanayis, University of Delaware, Newark, DE and Dr. Jinfang Liu, President & COO of Electron Energy Corporation, Landisville, PA.
The objectives of the workshop are to bring together engineers and scientists who work on permanent magnets and their applications to present their latest results and exchange ideas on a wide range of relevant topics including resources, raw materials, recycling, processing, fundamental and hard magnetic properties and applications of rare earth and future magnets.

To regsiter for this event, please follow the link below.

REPM 2021: The 26th International Workshop on Rare-Earth and Future Permanent Magnets and their Applications

The UKSG Annual Conference and Exhibition 2021 will be held online for the first time. The event is a major event in the scholarly communications calendar which attracts a large number of delegates each year from around the world - librarians, publishers, content providers, consultants and intermediaries. The conference combines high-quality plenary presentations, lightning talks and breakout sessions with virtual social events and a major online trade exhibition.



UKSG Annual Conference 2021

CPAF2021

Brought to you by **National Defense Industrial Association ([NDIA](https://www.google.com/url?q=https%3A%2F%2Fwww.ndia-sd.org%2F&sa=D&sntz=1&usg=AFQjCNHj_BQx1K4KVKKp1NTIsAltZOG9Vg))** and hosted by the **Battlespace Exploitation of Mixed Reality ([BEMR](https://cdn.dvidshub.net/media/video/2101/DOD_108139354/DOD_108139354-1024x576-1769k.mp4)) Lab**, at **Naval Information Warfare Center Pacific ([NIWC PAC](https://www.google.com/url?q=https%3A%2F%2Fwww.public.navy.mil%2Fnavwar%2FNIWC-Pacific%2FPages%2Fdefault.aspx&sa=D&sntz=1&usg=AFQjCNFNS160IEsocfaBrnCNB3u7NqLHPw))**, and welcoming academia, government/DOD, and industry leaders who support Industrial Augmented Reality/Virtual Reality by using the following technologies:   


* Augmented Reality (AR) / Virtual Reality (VR)

* 2D/3D displays

* mobile displays

* wearables

* any other "immersive" technology

### In all fields that touch:

* 3D modeling

* Manufacturing 

* Installations 

* Maintenance and Repairs

* Construction

* Logistics

* Additive manufacturing

* Training

The technical exchange is a two day event held online this year due to COVID-19. It will feature two days of oral presentations, demonstrations, and posters from leaders in AR/VR and other display technologies, including government briefings on active projects within the DOD. The technical exchange is unclassified and we welcome participation from the Navy, DoD as well as from industry and academia. 

The goals of TEAR 3M are:

*  To spread awareness of current immersive technology research relevant to Navy/DOD applications

*  To connect AR researchers with experts in DOD to learn about needs and requirements

*  To build and strengthen collaborations within the DOD research community

![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/0c13ac0ffa096b2353573964d8b89789.png)

TEAR 3M General Chairs:

Heidi Buck,  Director @ BEMR Lab, Naval Information Warfare Center, Pacific

Lauren Burnette, BEMR Lab, Naval Information Warfare Center, Pacific


For more information, contact Lauren and Heidi at:  tear3m@spawar.navy.mil

Questions regarding sponsorship/registration contact:  ndiasd@sandiegositepros.com

Conference host POC is Tammy Sanchez (NDIA):  tamsanch@cisco.com

TEAR 3M 2021

Welcome to Part I of the two-part, online workshop series on Aligning Global Engineering Graduate Program Priorities, funded by the National Science Foundation. The focus of Part I will be on the following questions:

Day 1: What is Global Engineering? What do global engineers need to know and know how to do?

Day 2: How do we achieve these learning objectives through graduate education? 

The goal of this workshop is to generate consensus on a Body of Knowledge within the emerging field of Global Engineering, working together to: (i) develop and align learning objectives and graduate curricula programming and (ii) strengthen collaboration amongst Global Engineering and other related programs in the United States. 

In preparation for the workshop, please watch the two plenary talks (12min, 19min) and review the following background documents: [North American Global Engineering Programs Summary](https://drive.google.com/file/d/12JHqNYsjdSavnrgk4GC28sy8g1Z0KAAc/view?usp=sharing); [Part I Workshop Participant Survey Summary](https://drive.google.com/file/d/1MROalnXDGBfTKlniq5d7rNd0IzvXFWy5/view?usp=sharing); and [Introduction to Bloom’s Taxonomy](https://drive.google.com/file/d/1i9XEjqhXvnKrUp5SLQ6WG9LWesBFMwOp/view?usp=sharing).

**Evan Thomas Plenary Video**[](https://)

[![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/478cee2d4c8e5daed1b09d2cb2285d46.png)](https://underline.io/events/72/sessions/2323/lecture/13272-evan-thomas-gew-plenary)

**Jack Colford Plenary Video**

[![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/0e9ea8f1579e4f113ced31d5a823e5d0.png)](https://underline.io/events/72/sessions/2328/lecture/13401-jack-colford-plenary)

Please view the [**[Workshop Participant List & Breakout Assignments]**](https://docs.google.com/spreadsheets/d/1CEZ8nnes5m1P2xrXLZLNyDZ9V2IOJ420J1erbCDsfY4/edit?usp=sharing)

**We use Gather.Town for the Breakout Sessions**

Please [click here](https://drive.google.com/file/d/1ewUD0IGC_-IgsHi2KZU1Y3F3zYWBPirV/view?usp=sharing) for written instructions on using Gather.Town


Aligning Global Engineering Graduate Program Priorities

Dear attendees,
we would like to remind you that this year's ISSCC has three events on Underline. If you wish to reach ISSCC Innovation [please click here](https://underline.io/events/74/reception). If you wish to reach ISSCC Exploration [please click here](https://underline.io/events/79/reception). Remember that you need to be registered for other events in order to have access.

[![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/8fdc308fdbe51b86add0a3713eff37fc.png)](https://underline.io/events/78/reception) [![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/b21b3a25c8bfec8565ed8b650d12f86b.png)](https://underline.io/events/74/reception) [![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/ded75d9813affa0f2ab1c2a2f35ce404.png)](https://underline.io/events/79/reception)


**The International Solid-State Circuits Conference** is the foremost global forum for presentation of advances in solid-state circuits and systems-on-a-chip. The Conference offers a unique opportunity for engineers working at the cutting edge of IC design and application to maintain technical currency, and to network with leading experts.

The ISSCC 2021 Technical Program is composed of three sub-conferences: Education, Innovations and Exploration. 
* Innovations consists of 205 outstanding technical papers including 4 plenary and 6 invited industry (“Highlighted Chip Releases”) and Special Events. 
* Education consists of twelve introductory tutorials, and a short course on PLLs, Clocking, and Clock Distribution. 
* Exploration consists of six expert-oriented forums.

The Conference will also feature a variety of special events, available to all attendees: 
* three panels on “What Technologies Will Shape the Future of Computing?”, “Going Remote: Challenges and Opportunities to Remote Learning, Work, and Collaboration”, and “Favorite Circuit Design and Testing Mistakes of Starting Engineers”, and 
* another technical session on “ICs in PandemICs”. In addition, we continue with the “Student Research Preview” and “Making a Career Choice”.

In addition, you will have access to two Demonstration Sessions, for both recorded and “live” sessions of devices described in nearly 50 selected papers.
Many networking options will be available, please use provided links to get access to networking lounges, speakers interview or book displays.

![](https://s3.eu-west-1.amazonaws.com/underline.prod/uploads/markdown_image/1/image/a3ba7e571668b614be66e31a2bbfd5b6.jpg)

<iframe width="560" height="315" src="https://www.youtube.com/embed/_SjbG9xnw3Q?controls=0" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

<iframe width=1280 height=706 frameborder="0" scrolling="no" src="https://screencast-o-matic.com/player/crnIjbSuyz?v=6&ff=1&title=0&controls=1" allowfullscreen="true"></iframe>

IEEE ISSCC Education

RWW consists of five related conferences that focus on the intersection between wireless communication theory, systems, circuits, and device technologies. This creates a unique forum for engineers to discuss various technologies for state-of-art wireless systems and their end-use applications.  ARFTG continues joining RWW2021 as a co-located conference.

RWW2021 will be held fully virtual on 17 – 22 January 2021.

IEEE Radio & Wireless Week (RWW)

ICPR 2020 is the premier international conference on Pattern Recognition, gathering leading researchers from around the world to present their latest achievements in Pattern Analysis and Recognition, applied Artificial Intelligence and Machine Learning. The topic of the year is " Putting Artificial intelligence to work on Patterns". 
For the first time, ICPR has employed a two-round review process similar to journal submissions, with new papers allowed to be submitted in either the first or second round and papers submitted in the first round and not accepted allowed to be revised and re-submitted for second round review. 

In this edition of the **ICPR 2020** Conference, there are 1407 Oral/Poster presentations organized in five Tracks: 
- Artificial Intelligence, Machine Learning for Pattern Analysis; 
- Biometrics, Human Analysis and Behavior Understanding; Computer Vision, Robotics and Intelligent Systems; 
- Document and Media Analysis; and 
- Image and Signal Processing. 

The **ICPR 2020** Main Conference program also includes 7 Keynotes from prominent world-renowned scientists, 8 Tutorials, 28 Demos, 8 Challenges, 2 Panels, as well as an Industry/Sponsor Exhibition. In addition, 41 Workshops on a range of important current and emerging research topics complete the ICPR 2020 program. 


ICPR 2020 MAIN CONFERENCE

ICPR 2020 Workshops & Tutorials

AAPT® is a strong professional physics science society dedicated to the pursuit of excellence in physical science education.

Winter and Summer National Meetings:
These National Meetings, held each winter and summer, are opportunities for members, colleagues, and future physicists from around the world to:

- participate in physics workshops
- meet and greet other physics educators
- form networks nationally and locally
- engage exhibitors and learn about the latest physics resources
- discuss innovations in teaching methods
- share the results of research about teaching and learning.

If you're not able to log in, you're most likely using **incorrect login credentials**. Double-check that you are using the same email and password that you used to register for the conference, submit your abstract or access your AAPT membership. There may also be **an issue with your AAPT meeting registration**. Please email aapt2021@underline.io and programs@aapt.org and we will work quickly to resolve this issue.

AAPT WM21

WACV is the premier international computer vision event comprising the main conference and several co-located workshops and tutorials. With its high quality and low cost, it provides an exceptional value for students, academics and industry researchers.



Wacv 2021

PErvasive Technologies Related to Assistive Environments

International Pediatric Stroke Organization

The IEEE Summer Topicals Meeting Series serves as an international forum to facilitate information exchange between various technical communities using or affected by rapidly growing areas of technology or “Hot Topics” related to the general field of Photonics. This intimate environment provides the opportunity to learn about emerging fields and to interact with the research and technology leaders. Historically, the conference hosts 4-7 individual topics and attracts 200+ participants with 30-50 participants attending each topic.

2021 IEEE Summer Topicals Meeting Series 

American Conference on Crystal Growth and Epitaxy

ComSciConCAN is Canada’s first national science communication workshop for graduate students. This unique professional development program aims to help the next generation of leaders in STEM fields develop the skills needed to communicate their research and ideas to their peers, experts in other fields, policy makers, and the general public.

ComSciConCAN looks to bring the longstanding and successful ComSciCon framework from the United States to Canada. Since its establishment in 2013 by graduate students at Harvard University and MIT, the training opportunities offered by ComSciCon have benefited over 500 graduate students across the United States and Canada. What was once a single annual national meeting has expanded to numerous local ComSciCon workshops across the United States. In collaboration with other alumni and invited experts, former ComSciCon attendees have published over 80 articles written at these workshops, initiated new science communication endeavours, and built lasting professional networks.


Due to the ongoing COVID-19 restrictions in place throughout Canada and beyond, CAN-ACN is excited to announce that the 14th Annual Meeting will be held as a virtual event, taking place from 23 – 25 August 2021. This decision was made to ensure the safety of our scientific community, as well as to enable maximum accessibility and engagement for those unable or unwilling to travel to Vancouver.

CAN-ACN 2021 

Join us in exploring emergent themes in PER as a scholarly endeavor and as a research community!  Since its inception, PER has served as a confluence of discipline, theory, research methodology, practice, and application.  How have these facets and the interplays among them been fruitful for physics education globally, regionally, locally, or otherwise?  And, how have our successes and interactions influenced the very nature of PER itself?
Following the lead of AAPT for the summer meeting, PERC 20/20 was virtual. Enjoy browsing the content.

PERC 2021

The UKSG Annual Conference and Exhibition is a major three-day event in the scholarly communications calendar. The conference attracts over 900 delegates from around the world including librarians, publishers, content providers, consultants and intermediaries. It takes place in the UK each spring and is open to all. 

Canadian Public Arts Funders (CPAF) is a network that brings together the federal, provincial and territorial arts councils, and equivalent public arts funders, to develop a shared conversation and understanding of common preoccupations and strategic development goals. This network provides an important space for support and strategic expertise; helping to create a more resilient arts funding sector as a whole.

Canadian Public Arts Funders

Technical Exchange on Augmented Reality for Military, Maintenance and Modeling (3M)

Part 1: Body of Knowledge and Learning Objectives

# **About ISSCC**

The International Solid-State Circuits Conference is the foremost global forum for presentation of advances in solid-state circuits and systems-on-a-chip. 
The Conference offers a unique opportunity for engineers working at the cutting edge of IC design and application to maintain technical currency, and to network with leading experts.

After our virtual paper selection process in October we are very happy to report, with strong submissions, we have put together an outstanding technical program together with an excellent set of educational programs. We are confident to bring you a high-quality conference as you all expect from ISSCC.

The International Solid-State Circuits Conference

Joint IEEE Radio and Wireless Week and ARFTG Microwave Measurement Week

**25th INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION**
MILAN, ITALY // JANUARY 10th TO 15th, 2021

ICPR 2020 is the premier international conference on Pattern Recognition, gathering leading researchers from around the world to present their latest achievements in Pattern Analysis and Recognition, applied Artificial Intelligence and Machine Learning. The topic of the year is " Putting Artificial intelligence to work on Patterns". 

For the first time, ICPR has employed a two-round review process similar to journal submissions, with new papers allowed to be submitted in either the first or second round and papers submitted in the first round and not accepted allowed to be revised and re-submitted for second round review. 

In this edition of the ICPR Conference, there are 1407 Oral/Poster presentations organized in five Tracks: Artificial Intelligence, Machine Learning for Pattern Analysis; Biometrics, Human Analysis and Behavior Understanding; Computer Vision, Robotics and Intelligent Systems; Document and Media Analysis; and Image and Signal Processing. The ICPR2020 Main Conference program also includes 7 Keynotes from prominent world-renowned scientists, 8 Tutorials, 28 Demos, 8 Challenges, 2 Panels, as well as an Industry/Sponsor Exhibition. In addition, 41 Workshops on a range of important current and emerging research topics complete the ICPR2020 program. 


25th International Conference on Pattern Recognition

**25th International Conference on Pattern Recognition.**
25th INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION MILAN, ITALY // JANUARY 10th TO 15th, 2021

ICPR 2020 is the premier international conference on Pattern Recognition, gathering leading researchers from around the world to present their latest achievements in Pattern Analysis and Recognition, applied Artificial Intelligence and Machine Learning. The topic of the year is " Putting Artificial intelligence to work on Patterns".

For the first time, ICPR has employed a two-round review process similar to journal submissions, with new papers allowed to be submitted in either the first or second round and papers submitted in the first round and not accepted allowed to be revised and re-submitted for second round review.

In this edition of the ICPR Conference, there are 1407 Oral/Poster presentations organized in five Tracks: Artificial Intelligence, Machine Learning for Pattern Analysis; Biometrics, Human Analysis and Behavior Understanding; Computer Vision, Robotics and Intelligent Systems; Document and Media Analysis; and Image and Signal Processing. The ICPR2020 Main Conference program also includes 7 Keynotes from prominent world-renowned scientists, 8 Tutorials, 28 Demos, 8 Challenges, 2 Panels, as well as an Industry/Sponsor Exhibition. In addition, 41 Workshops on a range of important current and emerging research topics complete the ICPR2020 program.

2021 AAPT Virtual Winter Meeting

English (automatic)

https://d3smihljt9218e.cloudfront.net/transcript/13574/subtitle/354286b41241bc886e9f7c74ae6178ba.vtt

**Read the paper on the folowing link:**

https://www.aclweb.org/anthology/2021.naacl-main.103/

**Abstract:**

Recent neural text-to-SQL models can effectively translate natural language questions to corresponding SQL queries on unseen databases. Working mostly on the Spider dataset, researchers have proposed increasingly sophisticated solutions to the problem. Contrary to this trend, in this paper we focus on simplifications. We begin by building DuoRAT, a re-implementation of the state-of-the-art RAT-SQL model that unlike RAT-SQL is using only relation-aware or vanilla transformers as the building blocks. We perform several ablation experiments using DuoRAT as the baseline model. Our experiments confirm the usefulness of some techniques and point out the redundancy of others, including structural SQL features and features that link the question with the schema


DuoRAT: Towards Simpler Text-to-SQL Models

**Read the paper on the folowing link:**

https://www.aclweb.org/anthology/2021.naacl-main.102/

**Abstract:**

Negation is a core construction in natural language. Despite being very successful on many tasks, state-of-the-art pre-trained language models often handle negation incorrectly. To improve language models in this regard, we propose to augment the language modeling objective with an unlikelihood objective that is based on negated generic sentences from a raw text corpus. By training BERT with the resulting combined objective we reduce the mean top 1 error rate to 4% on the negated LAMA dataset. We also see some improvements on the negated NLI benchmarks.


Understanding by Understanding Not: Modeling Negation in Language Models

**Read the paper on the folowing link:**

https://www.aclweb.org/anthology/2021.naacl-srw.12/

**Abstract:**



Shuffled-token Detection for Refining Pre-trained RoBERTa

**Read the paper on the folowing link:**

https://www.aclweb.org/anthology/2021.naacl-main.104/

**Abstract:**

Natural Language Inference (NLI) has garnered significant attention in recent years; however, the promise of applying NLI breakthroughs to other downstream NLP tasks has remained unfulfilled. In this work, we use the multiple-choice reading comprehension (MCRC) and checking factual correctness of textual summarization (CFCS) tasks to investigate potential reasons for this. Our findings show that: (1) the relatively shorter length of premises in traditional NLI datasets is the primary challenge prohibiting usage in downstream applications (which do better with longer contexts); (2) this challenge can be addressed by automatically converting resource-rich reading comprehension datasets into longer-premise NLI datasets; and (3) models trained on the converted, longer-premise datasets outperform those trained using shortpremise traditional NLI datasets on downstream tasks primarily due to the difference in premise lengths.


Looking Beyond Sentence-Level Natural Language Inference for Question Answering and Text Summarization

This paper presents our contribution to the ProfNER shared task. Our work focused on evaluating different pre-trained word embedding representations suitable for the task. We further explored combinations of embeddings in order to improve the overall results.

Assessing multiple word embeddings for named entity recognition of professions and occupations in health-related social media

**Read the paper on the folowing link:**

https://www.aclweb.org/anthology/2021.naacl-srw.13/

**Abstract:**



Morphology-Aware Meta-Embeddings for Tamil

**Read the paper on the folowing link:**

https://www.aclweb.org/anthology/2021.naacl-main.105/

**Abstract:**

Learning to capture text-table alignment is essential for tasks like text-to-SQL. A model needs to correctly recognize natural language references to columns and values and to
ground them in the given database schema. In this paper, we present a novel weakly supervised Structure-Grounded pretraining framework (STRUG) for text-to-SQL that can effectively learn to capture text-table alignment based on a parallel text-table corpus. We identify a set of novel pretraining tasks: column grounding, value grounding and column value mapping, and leverage them to pretrain a text-table encoder. Additionally, to evaluate different methods under more realistic text-table alignment settings, we create a new evaluation set Spider-Realistic based on Spider dev set with explicit mentions of column names removed, and adopt eight existing text-to-SQL datasets for cross-database evaluation. STRUG brings significant improvement over BERT-LARGE in all settings. Compared with existing pretraining methods such as GRAPPA, STRUG achieves similar performance on Spider, and outperforms all baselines on more realistic sets.


Structure-Grounded Pretraining for Text-to-SQL

**Read the paper on the folowing link:**

https://www.aclweb.org/anthology/2021.naacl-main.90/

**Abstract:**

While traditional corpus-level evaluation metrics for machine translation (MT) correlate well with fluency, they struggle to reflect adequacy. Model-based MT metrics trained on segment-level human judgments have emerged as an attractive replacement due to strong correlation results. These models, however, require potentially expensive re-training for new domains and languages. Furthermore, their decisions are inherently non-transparent and appear to reflect unwelcome biases. We explore the simple type-based classifier metric, MacroF1, and study its applicability to MT evaluation. We find that MacroF1 is competitive on direct assessment, and outperforms others in indicating downstream cross-lingual information retrieval task performance. Further, we show that MacroF1 can be used to effectively compare supervised and unsupervised neural machine translation, and reveal significant qualitative differences in the methods' outputs. 


Macro-Average: Rare Types Are Important Too

The paper researches the problem of drug adverse effect detection in texts of social media. We describe the development of such classification system for Russian tweets. To increase the train dataset we apply a couple of augmentation techniques and analyze their effect in comparison with similar systems presented at 2021 years' SMM4H Workshop.

Text Augmentation Techniques in Drug Adverse Effect Detection Task

**Read the paper on the folowing link:**

https://www.aclweb.org/anthology/2021.naacl-main.91/

**Abstract:**

Reference-free evaluation has the potential to make machine translation evaluation substantially more scalable, allowing us to pivot easily to new languages or domains. It has been recently shown that the probabilities given by a large, multilingual model can achieve state-of-the-art results when used as a reference-free metric. We experiment with various modifications to this model and demonstrate that by scaling it up we can match the performance of BLEU. We analyze various potential weaknesses of the approach and find that it is surprisingly robust and likely to offer reasonable performance across a broad spectrum of domains and different system qualities.


Disentangling Semantics and Syntax in Sentence Embeddings with Pre-trained Language Models

James Y. Huang

Kai-Wei Chang

Kuan-Hao Huang

Disentangling Semantics and Syntax in Sentence Embeddings with Pre-trained Language Models

James Y. Huang

Kai-Wei Chang

Kuan-Hao Huang

similar conferences

NAACL 2021

WACV 2021

WASH21

2021EL

Petra2021

IPSO2021

SUM2021

ACCGE-22/OMVPE-20

ComSciConCAN2021

CAN-ACN

PERC Meeting 2021

REPM 2021

UKSG Annual 2021

CPAF

TEAR 3M 2021

Aligning Global Engineering

IEEE ISSCC - Education

RWW_ARFTG 2021

ICPR 2021

ICPR 2021 Workshops

AAPT WM21

DuoRAT: Towards Simpler Text-to-SQL Models

similar conferences

NAACL 2021

WACV 2021

WASH21

2021EL

Petra2021

IPSO2021

SUM2021

ACCGE-22/OMVPE-20

ComSciConCAN2021

CAN-ACN

PERC Meeting 2021

REPM 2021

UKSG Annual 2021

CPAF

TEAR 3M 2021

Aligning Global Engineering

IEEE ISSCC - Education

RWW_ARFTG 2021

ICPR 2021

ICPR 2021 Workshops

AAPT WM21