Publications & Patents

Conference Publications

LOCATE: Self-supervised Object Discovery via Flow-guided Graph-cut and Bootstrapped Self-training

S Singh, S Deshmukh, M Sarkar, B Krishnamurthy. Accepted at British Machine Vision Conference (BMVC), 2023.

Explaining RL Decisions using Trajectories

S Deshmukh, A Dasgupta, C Agarwal, N Jiang, B Krishnamurthy, G Theocharous, J Subramanian. Published at International Conference on Learning Representations (ICLR), 2023.

One-Shot Doc Snippet Detection: Powering Search in Document Beyond Text

A Java*, S Deshmukh*, M Aggarwal, S Jandial, M Sarkar, B Krishnamurthy (*Equal contribution). Published at IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023.

Explain Your Move: Understanding Agent Actions Using Focused Feature Saliency

P Gupta, N Puri, S Verma, D Kayastha, S Deshmukh, B Krishnamurthy, S Singh. Published at International Conference on Learning Representations (ICLR), 2020.

Workshop Publications

Counterfactual Explanation Policies in RL

S Deshmukh, Srivatsan R, S Vijay, J Subramanian, C Agarwal. Published at ICML Workshop on Counterfactuals in Minds & Machines, 2023.

Gatha: Relational Loss for enhancing text-based style transfer

S Jandial, S Deshmukh, A Java, S Shahid, B Krishnamurthy. Published (Oral) at CVPR Workshop on Computer Vision for Fashion, Art, and Design, 2023.

FODVid: Flow-guided Object Discovery in Videos

S Singh, S Deshmukh, M Sarkar, R Jain, M Hemani, B Krishnamurthy. Published at CVPR Workshop on Learning with Limited Labelled Data for Image and Video Understanding, 2023.

Trajectory-based Explainability Framework for Offline RL

S Deshmukh, A Dasgupta, C Agarwal, N Jiang, B Krishnamurthy, G Theocharous, J Subramanian. Published at NeurIPS Offline RL Workshop, 2022.

Patents

Novel Self-supervised Object Discovery in Videos

Attorney Docket No. P12351-US, 2023

Account Executive Actionable Digest

Attorney Docket No. P12314-US, 2023

A Framework for Leveraging LLM Models and RL in Marketing Decision Making

Attorney Docket No. P12223-US, 2023

Video Object Segmentation through Flow-guided Graph-cut

Attorney Docket No. P12170-US, 2023

Novel method to propagate personalized error corrections across Forms corpus

Attorney Docket No. P12004-US, 2022

Novel Trajectory-based Explainability Framework for RL-based Decision Making

Attorney Docket No. P11853-US, 2022

Forms Similarity Matching Framework for Enhancing RnC tool in AEM Forms

Attorney Docket No. P11882-US, 2022

A Novel Multimodal One-Shot Detection Approach for Document Snippet Search

Attorney Docket No. P11686-US, 2022

Semantic Noise based Soft Label Regularization for Distilling Model Knowledge

Attorney Docket No. P11539-US, 2022

Novel method to simplify data points for easier understanding of neural networks

Attorney Docket No. P11364-US, 2022

Novel Method and Apparatus to Control Diffusion Model Image Generation

Attorney Docket No. P11343 -US, 2022

Refining Element Associations for Form Structure Extraction

Attorney Docket No. P10768-US, 2021

Customer Journey Management Using Machine Learning

Attorney Docket No. P10405-US, 2021