“I worked with Emad in FUSE Labs (Microsoft Research) for 4 years. He is a dynamic developer and a fast learner, able to move between technologies quickly. He's also a pleasure to work with - always friendly and communicates well. He'd be a win for any team.”
Experience & Education
Licenses & Certifications
Volunteer Experience
-
Contributor
Wikimedia Foundation
- Present 10 years
Education
Made code contribution to the open source MediaWiki Core code base.
-
IT Director and Webmaster
CIE'38 Conference
- 6 months
Science and Technology
I was responsible for all the IT for the conference, including the website, paper submission system, review system, scheduling system, electronic proceedings and workshop management.
-
Event Planner and Coordinator
SOS Music Festival
- 4 months
Arts and Culture
-
President
Brookside Villas HOA
- Present 8 years 8 months
Publications
-
DeeperDive: The Unreasonable Effectiveness of Weak Supervision in Document Understanding
ACM SIGKDD 2022
Weak supervision has been applied to various Natural Language Understanding tasks in recent years. Due to technical challenges with scaling weak supervision to work on long-form documents, spanning up to hundreds of pages, applications in the document understanding space have been limited. At Lexion, we built a weak supervision-based system tailored for long-form (10-200 pages long) PDF documents.
We use this platform for building dozens of language understanding models and have applied it…Weak supervision has been applied to various Natural Language Understanding tasks in recent years. Due to technical challenges with scaling weak supervision to work on long-form documents, spanning up to hundreds of pages, applications in the document understanding space have been limited. At Lexion, we built a weak supervision-based system tailored for long-form (10-200 pages long) PDF documents.
We use this platform for building dozens of language understanding models and have applied it successfully to various domains, from commercial agreements to corporate formation documents.
In this paper, we demonstrate the effectiveness of supervised learning with weak supervision in a situation with limited time, workforce, and training data. We built 8 high quality machine learning models in the span of one week, with the help of a small team of just 3 annotators working with a dataset of under 300 documents. We share some details about our overall architecture, how we utilize weak supervision, and what results we are able to achieve. We also include the dataset for researchers who would like to experiment with alternate approaches or refine ours.
Furthermore, we shed some light on the additional complexities that arise when working with poorly scanned long-form documents in PDF format, and some of the techniques that help us achieve state-of-the-art performance on such data.Other authorsSee publication -
The Law of Large Documents: Understanding the Structure of Legal Contracts Using Visual Cues
ACM SIGKDD 2021: Special Interest Group on Knowledge Discovery and Data Mining
Large, pre-trained transformer models like BERT have achieved state-of-the-art results on document understanding tasks, but most implementations can only consider 512 tokens at a time. For many real-world applications, documents can be much longer, and the segmentation strategies typically used on longer documents miss out on document structure and contextual information, hurting their results on downstream tasks. In our work on legal agreements, we find that visual cues such as layout, style…
Large, pre-trained transformer models like BERT have achieved state-of-the-art results on document understanding tasks, but most implementations can only consider 512 tokens at a time. For many real-world applications, documents can be much longer, and the segmentation strategies typically used on longer documents miss out on document structure and contextual information, hurting their results on downstream tasks. In our work on legal agreements, we find that visual cues such as layout, style, and placement of text in a document are strong features that are crucial to achieving an acceptable level of accuracy on long documents. We measure the impact of incorporating such visual cues, obtained via computer vision methods, on the accuracy of document understanding tasks including document segmentation, entity extraction, and attribute classification. Our method of segmenting documents based on structural metadata out-performs existing methods on four long-document understanding tasks as measured on the Contract Understanding Atticus Dataset.
Other authorsSee publication -
BERT Goes to Law School: Quantifying the Competitive Advantage of Access to Large Legal Corpora in Contract Understanding
Thirty-third Conference on Neural Information Processing Systems (NeurIPS 2019)
Fine-tuning language models, such as BERT, on domain specific corpora has proven to be valuable in domains like scientific papers and biomedical text. In this paper, we show that fine-tuning BERT on legal documents similarly provides valuable improvements on NLP tasks in the legal domain. Demonstrating this outcome is significant for analyzing commercial agreements, because obtaining large legal corpora is challenging due to their confidential nature. As such, we show that having access to…
Fine-tuning language models, such as BERT, on domain specific corpora has proven to be valuable in domains like scientific papers and biomedical text. In this paper, we show that fine-tuning BERT on legal documents similarly provides valuable improvements on NLP tasks in the legal domain. Demonstrating this outcome is significant for analyzing commercial agreements, because obtaining large legal corpora is challenging due to their confidential nature. As such, we show that having access to large legal corpora is a competitive advantage for commercial applications, and academic research on analyzing contracts.
Other authorsSee publication -
Calendar.help: Designing a Workflow-Based Scheduling Agent with Humans in the Loop
ACM CHI Conference on Human Factors in Computing Systems
Although information workers may complain about meetings, they are an essential part of their work life. Consequently, busy people spend a significant amount of time scheduling meetings. We present Calendar.help, a system that provides fast, efficient scheduling through structured workflows. Users interact with the system via email, delegating their scheduling needs to the system as if it were a human personal assistant. Common scheduling scenarios are broken down using well-defined workflows…
Although information workers may complain about meetings, they are an essential part of their work life. Consequently, busy people spend a significant amount of time scheduling meetings. We present Calendar.help, a system that provides fast, efficient scheduling through structured workflows. Users interact with the system via email, delegating their scheduling needs to the system as if it were a human personal assistant. Common scheduling scenarios are broken down using well-defined workflows and completed as a series of microtasks that are automated when possible and executed by a human otherwise. Unusual scenarios fall back to a trained human assistant who executes them as unstructured macrotasks. We describe the iterative approach we used to develop Calendar.help, and share the lessons learned from scheduling thousands of meetings during a year of real-world deployments. Our findings provide insight into how complex information tasks can be broken down into repeatable components that can be executed efficiently to improve productivity.
Other authorsSee publication -
Enhancing Cortana User Experience Using Machine Learning
Stanford University
Voice enabled personal assistants like Microsoft Cortana are becoming better every day. As a result more users are relying on such software to accomplish more tasks. While these applications are significantly improving due to great advancements in the underlying technologies, there are still shortcomings in their performance resulting in a class of user queries that such assistants cannot yet handle with satisfactory results.
We analyze the data from millions of user queries, and build a…Voice enabled personal assistants like Microsoft Cortana are becoming better every day. As a result more users are relying on such software to accomplish more tasks. While these applications are significantly improving due to great advancements in the underlying technologies, there are still shortcomings in their performance resulting in a class of user queries that such assistants cannot yet handle with satisfactory results.
We analyze the data from millions of user queries, and build a machine learning system capable of classifying user queries into two classes; a class of queries that are addressable by Cortana with high user satisfaction, and a class of queries that are not. We then use unsupervised learning to cluster similar queries and assign them to human assistants who can complement Cortana functionality.
Other authorsSee publication -
Quantifying the Impact of GPU Specific Optimizations: An Experimental Study on a Weather Forecasting Application
ACM PACT, Austria
The highly parallel nature and computational power of GPUs were utilized to enhance performance by porting computationally intensive modules of the Weather Research and Forecasting's (WRF) model from single threaded FORTRAN to multithreaded CUDA code. GPU specific optimizations were then made, and various testing and benchmarking tools were used to validate this enhancement. The resulting code has been integrated as part of the WRF's official V3.2 release; it has also been published on the…
The highly parallel nature and computational power of GPUs were utilized to enhance performance by porting computationally intensive modules of the Weather Research and Forecasting's (WRF) model from single threaded FORTRAN to multithreaded CUDA code. GPU specific optimizations were then made, and various testing and benchmarking tools were used to validate this enhancement. The resulting code has been integrated as part of the WRF's official V3.2 release; it has also been published on the University Corporation for Atmospheric Research's proceedings. The accelerated version was also deployed with a friendly UI, offering local weather forecasts.
Other authorsSee publication
Patents
-
Workflow Engine Design as a Dynamic Directed Graph
Filed US US 363075.01
-
Assisted Micro-tasks that Combine Human and Machine Intelligence in a Virtual Assistant
Filed US US 362191.01
-
Agent system combining machine learning, micro tasks, and macro tasks
Filed US 358994.01
Agent systems need to be accurate, flexible, scalable, and inexpensive. Systems based fully on machine learning are scalable and inexpensive, but they often fail to be accurate and flexible. Micro tasking systems are also prone to failing at accuracy and flexibility. Macro tasking systems are flexible and accurate, but they are expensive and don't scale well. What we have invented is an architecture that can take advantage of machine learning and micro task work, but provides a mechanism to…
Agent systems need to be accurate, flexible, scalable, and inexpensive. Systems based fully on machine learning are scalable and inexpensive, but they often fail to be accurate and flexible. Micro tasking systems are also prone to failing at accuracy and flexibility. Macro tasking systems are flexible and accurate, but they are expensive and don't scale well. What we have invented is an architecture that can take advantage of machine learning and micro task work, but provides a mechanism to transfer a task to a macro task worker when that is necessary to solve the problem properly.
Other inventors
Projects
-
calendar.help
Calendar.help does one thing really well: schedule your meetings.
With the speed of artificial intelligence and the personal touch of a human assistant, Calendar.help takes care of business.Other creatorsSee project -
Workflow Engine for long running processses
- Present
This Library allows for creating long running, stateful and versioned workflows for breaking complex workflows into smaller ones that can be run synchronously or asynchronously over short or long periods of time.
-
HereHere NYC
HereHere NYC is a research project that explores creating compelling stories with data to engage larger communities, inventing daily rituals for connecting to the hyper-local and using characterization as a tool to drive data engagement.
Other creatorsSee project
Honors & Awards
-
40 under 40
Puget Sound Business Journal
The Puget Sound Business Journal is once again putting a spotlight on those who have established themselves as creative, innovative industry leaders before reaching the age of 40.
https://www.bizjournals.com/seattle/news/2021/07/29/psbjs-40-under-40-class-of-2021-revealed-part-1.html -
Bot Framework //build 2017 release
Microsoft
-
Ship It Award - Kodu
Microsoft
-
Ship It Award - Bing Neon Release
Microsoft
-
Achieving G.R.E.A.T Results
Microsoft
-
Ship It Award - Bing Metallica Release
Microsoft Corporation
-
Ship It Award - Bing Mt. View Release
Microsoft Corporation
-
Gold Star Award for Engineering Excellence and Enabling Others
Microsoft
-
Ship It Award - Bing Denver Release
Microsoft Corporation
-
Computer Scientific Society Award
Arab Academy of Science and Technology
-
Ship It Award - Bing Boston Release
Microsoft
-
Best Egyptian Young Programmer
Egyptian Ministry of Communication and Information
Organizations
-
Seattle City Rotaract
Member
- Present -
ACM
Professional Member
- Present
Recommendations received
1 person has recommended Emad
Join now to viewOther similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Emad Elwany
-
Emad Elwany
Director at Theater!
-
emad elwany
general manager at El-Zahraa for plastic
-
Emad Elwany
--
3 others named Emad Elwany are on LinkedIn
See others named Emad Elwany