Solving Price Per Unit Problem Around the World: Formulating Fact Extraction as Question Answering

no code implementations • 12 Apr 2022 • Tarik Arici, Kushal Kumar, Hayreddin Çeker, Anoop S V K K Saladi, Ismail Tutar

Our model architecture consists of two subnetworks for the two subtasks: a classifier to predict UoM type (or the question) and an extractor to extract the relevant quantities.

named-entity-recognition Named Entity Recognition +2

MLIM: Vision-and-Language Model Pre-training with Masked Language and Image Modeling

no code implementations • 24 Sep 2021 • Tarik Arici, Mehmet Saygin Seyfioglu, Tal Neiman, Yi Xu, Son Train, Trishul Chilimbi, Belinda Zeng, Ismail Tutar

Vision-and-Language Pre-training (VLP) improves model performance for downstream tasks that require image and text inputs.

Image Reconstruction Language Modelling +1

MULTI-SPAN QUESTION ANSWERING USING SPAN-IMAGE NETWORK

no code implementations • 1 Jan 2021 • Tarik Arici, Hayreddin Ceker, Ismail Baha Tutar

Question-answering (QA) models aim to find an answer given a question and con- text.

Question Answering valid

Associative Adversarial Networks

no code implementations • 18 Nov 2016 • Tarik Arici, Asli Celikyilmaz

In this work, we use Restricted Boltzmann Machines (RBMs) as a higher-level associative memory and learn the probability distribution for the high-level features generated by D. The associative memory samples its underlying probability distribution and G learns how to map these samples to data space.

Generative Adversarial Network