added RAMEN

Author: aneesh-shetye

natural_language_processing/README.md

@@ -4,3 +4,4 @@
 |:---------------------------------------------------------------------------------------------------------:|:---------------------------------------------------:|:----------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------:|
 | [Sequence to Sequence Learning with Neural Networks](https://arxiv.org/abs/1409.3215)                     | [HackMD](https://hackmd.io/@photon-dodo/HyrN0wjkv)  | [Rishika](https://https://github.com/rishika2110) [Khurshed](https://https://github.com/GlazeDonuts) | This paper presents a novel architecture and paradigm for Neural Machine Translation. |
 | [Neural Machine Translation by jointly learning to align and translate]( https://arxiv.org/abs/1409.0473) | [HackMD]( https://hackmd.io/@photon-dodo/HJfefAQbP) | [Rishika](https://https://github.com/rishika2110) [Khurshed](https://https://github.com/GlazeDonuts) | This paper introduces a novel attention based approach for translation.               |
+| [Answer Them All! Toward Universal Visual Question Answering Models](https://arxiv.org/abs/1903.00366)     | [Notion](https://phrygian-macaroni-e3b.notion.site/Answer-Them-All-RAMEN-d441cd8797984474baeba5ce4176956d) | [Aneesh](https://sites.google.com/view/aneesh-shetye/home) | This paper tries to resolve the disparity in performance of previous Visual Question Answering (VQA) architectures on synthetic and natural datasets by introducing a novel architecture. |