Quality Estimation without Human-labeled Data

April 21, 2021 By: Yi-Lin Tuan, Ahmed El-Kishky, Adithya Renduchintala, Vishrav Chaudhary, Francisco Guzmán, Lucia Specia Abstract Quality estimation aims to measure the quality of translated content without access to a reference translation. This is crucial for machine translation systems in real-world scenarios where high-quality translation is needed. While many approaches exist for quality estimation, they are based on supervised machine learning requiring costly human labelled data. As an alternative, we propose a technique that does not rely on examples […]

Read more

MLQA: Evaluating Cross-lingual Extractive Question Answering

Abstract Question answering (QA) models have shown rapid progress enabled by the availability of large, high-quality benchmark datasets. Such annotated datasets are difficult and costly to collect, and rarely exist in languages other than English, making building QA systems that work well in other languages challenging. In order to develop such systems, it is crucial to invest in high quality multilingual evaluation benchmarks to measure progress. We present MLQA, a multi-way aligned extractive QA evaluation benchmark intended to spur research […]

Read more

Machine Learning Automation using EvalML Library

This article was published as a part of the Data Science Blogathon Introduction Machine Learning is one of the fastest-growing technology in the modern era. New innovations in the field of ML and AI are made each and every day which supports the world to leap forward. Earlier for a person entering into the ML field finds it difficult to create accurate machine learning models, but now AutoML Libraries are created which helps the beginners to create an accurate model with […]

Read more

My most amazing Makefile for CL papers

Automation of stuff that does not need to be automated at all is one of my most favorite procrastination activities. As an experienced (and most of the time unsuccessful) submitter to conferences organized by ACL (ACL, NAACL, EACL, EMNLP), I spent a lot of procrastinating time improving the Makefile compiling the papers. Here are few commented snippets from the Makefiles. Hopefully, someone finds that useful. The normal LaTeX stuff I compile the paper using latexmk. main.pdf: $(FILES) latexmk -pdflatex=”$(LATEX) %O […]

Read more

Anticipate, absorb, and adapt—introducing the societal resilience research agenda

The genetic sequence for COVID-19 was first published in January 2020. Before the end of the year, new vaccines—which typically take five to ten years to develop—were approved for emergency use in multiple nations. This astounding achievement is emblematic of a “reset” in the relationship between science and society. Faced with a frightening health crisis, the world called on scientists and researchers  

Read more

Issue #131 – Measuring and Increasing Context Usage in Context-Aware NMT

20 May21 Issue #131 – Measuring and Increasing Context Usage in Context-Aware NMT in Model improvement, The Neural MT Weekly Author: Dr. Patrik Lambert, Senior Machine Translation Scientist @ Iconic Introduction Context-aware neural machine translation (NMT) is a topic which has often been covered in this blog, for its application to domain adaptation or document-level NMT (see issues #15, #31, #34, #39, #98, #128). However, most papers on context-aware NMT present approaches which have the ability to leverage context information, […]

Read more

A video games engine for python

PursuedPyBear PursuedPyBear, also known as ppb, exists to be an educational resource. Most obviously used to teach computer science, it can be a useful tool for any topic that a simulation can be helpful. A Game Engine At its core, ppb provides a number of features that make it perfect for video games. The GameEngine itself provides a pluggable subsystem architecture where adding new features is as simple as subclassing and extending System. Additionally, it contains a state stack of […]

Read more

A tool for automatically generating 3D printable STLs from freely available lidar scan data

mini-map-maker A tool for automatically generating 3D printable STLs from freely available lidar scan data. To use this script, go to the USGS LidarExplorer https://prd-tnm.s3.amazonaws.com/LidarExplorer/index.html#/ Select an area, and then click the “Download list” button under “Lidar within AOI”This should give you a file called downloadlist.txt. Simply place this text file in the same directory as the script,and then run the script (convert.py, or convert.exe if you don’t want to deal with Python). By default,mini-map-maker will generate an STL file […]

Read more

A Discord API Wrapper for Userbots/Selfbots written in Python

DisCum A simple, easy to use, non-restrictive, synchronous Discord API Wrapper for Selfbots/Userbots written in Python. Discum is a Discord self/userbot api wrapper (in case you didn’t know, self/userbotting = automating a user account). Whenever you login to discord, your client communicates with Discord’s servers using Discord’s http api (http(s) requests) and gateway server (websockets). Discum allows you have this communication with Discord using python. The main difference between Discum and other Discord api wrapper libraries (like discord.py) is that […]

Read more

Marshall python objects to and from JSON

Pymarshaler Pymarshaler allows you to marshal and unmarshal any python object directly to and from a JSON formatted string. Pymarshaler takes advantage of python’s new typing support. By reading class init param types, we are able to walk down nested JSON structures and assign appropriate values. Basic Usage Declare a class with typing information class Test: def __init__(self, name: str): self.name = name That’s it! We can now marshal, and more importantly, unmarshal this object to and from JSON. from […]

Read more
1 690 691 692 693 694 985