Self-supervised Multi-view Disentanglement for Expansion of Visual Collections

no code implementations • 4 Feb 2023 • Nihal Jain, Praneetha Vaddamanu, Paridhi Maheshwari, Vishwa Vinay, Kuldeep Kulkarni

In this work, we consider the setting where a query for similar images is derived from a collection of images.

Disentanglement Image Retrieval +2

ContraCLM: Contrastive Learning For Causal Language Model

no code implementations • 3 Oct 2022 • Nihal Jain, Dejiao Zhang, Wasi Uddin Ahmad, Zijian Wang, Feng Nan, Xiaopeng Li, Ming Tan, Ramesh Nallapati, Baishakhi Ray, Parminder Bhatia, Xiaofei Ma, Bing Xiang

Specifically, we attain $44\%$ relative improvement on the Semantic Textual Similarity tasks and $34\%$ on Code-to-Code Search tasks.

Code Generation Code Search +4

MultiViz: Towards Visualizing and Understanding Multimodal Models

1 code implementation • 30 Jun 2022 • Paul Pu Liang, Yiwei Lyu, Gunjan Chhablani, Nihal Jain, Zihao Deng, Xingbo Wang, Louis-Philippe Morency, Ruslan Salakhutdinov

How can we visualize the internal modeling of multimodal interactions in these models?

Generating Compositional Color Representations from Text

no code implementations • 22 Sep 2021 • Paridhi Maheshwari, Nihal Jain, Praneetha Vaddamanu, Dhananjay Raut, Shraiysh Vaishay, Vishwa Vinay

While this dataset is specialized for our investigations on color, the method can be extended to other visual dimensions where composition is of interest.

Attribute Contrastive Learning +3