Publications

I am interested in wide range of topics in AI, currently very excited about the developments in Foundation Models and Reinforcement Learning. Part of my research in the future looks at building efficient Foundation Models and at the interface of Language Models and Game Theory

Frugal lms trained to invoke symbolic solvers achieve parameter-efficient arithmetic reasoning

https://ojs.aaai.org/index.php/AAAI/article/view/29750

Venue: AAAI, Oral

Year: 2024

Notwithstanding great recent enthusiasm about LLMs as allpurpose problem solvers, practitioners appreciate that LLMs work best when limited in their role to act as a glue between tools specialized to non-linguistic tasks like logic, arithmetic, or structured information retrieval. In response, LLMs are steadily getting better at invoking tools. Here, through the design of a new system, SYRELM, we explore a synergy between symbolic and numeric reasoning that has been established in middle-school pedagogy for a while, but not yet commonplace with LLMs.

Small Language Models Fine-tuned to Coordinate Larger Language Models improve Complex Reasoning

https://aclanthology.org/2023.emnlp-main.225/

Venue: EMNLP

Year: 2023

We challenged the design of ever-larger monolithic LLMs as homogeneous network structures, where diverse aspects of problem decomposition and solution are stored in a tangled and opaque manner. The formidable general-purpose problem-solving capabilities of LLMs are exceedingly resource-hungry, dependent on immense data engineering. Inspired by brain science, we took a first step toward heterogeneity — let two different LLMs evolve independently and adapt to their roles of decomposing and solving complex reasoning problems. Through extensive experiments on several benchmarks, we showed that such a heterogeneous network can match or exceed some of the largest contemporary LLMs, at a much smaller parameter count.

D-LORD: DYSL-AI database for low-resolution disguised face recognition

https://ieeexplore.ieee.org/abstract/document/10224654

Venue: IEEE Transactions on Biometrics, Behavior, and Identity Science 

Year: 2024

Face recognition in a low-resolution video stream captured from a surveillance camera is a challenging problem. The problem becomes even more complicated when the subjects appearing in the video wear disguise artifacts to hide their identity or try to impersonate someone. The lack of labeled datasets restricts the current research on low-resolution face recognition systems under disguise. With this paper, we propose a large-scale database, D-LORD, that will facilitate the research on face recognition.

Seg-dgdnet: Segmentation based disguise guided dropout network for low resolution face recognition

https://ieeexplore.ieee.org/abstract/document/10163869

Term: IEEE Journal of Selected Topics in Signal Processing

Offered: 2023

Include the official course description from your institution. Include any other relevant information needed for students or others viewing to understand major themes, prerequisites or if classes are exclusive for specific major subjects or degree tracks.

IDD-AW: a benchmark for safe and robust segmentation of drive scenes in unstructured traffic and adverse weather

https://openaccess.thecvf.com/content/WACV2024/papers/Shaik_IDD-AW_A_Benchmark_for_Safe_and_Robust_Segmentation_of_Drive_WACV_2024_paper.pdf

Term: WACV

Offered: 2024

Large-scale deployment of fully autonomous vehicles requires a very high degree of robustness to unstructured traffic, weather conditions, and should prevent unsafe mispredictions. While there are several datasets and benchmarks focusing on segmentation for drive scenes, they are not specifically focused on safety and robustness issues. We introduce the IDD-AW dataset, which provides 5000 pairs of high-quality images with pixel-level annotations, captured under rain, fog, low light, and snow in unstructured driving conditions. As compared to other adverse weather datasets, we provide i.) more annotated images, ii.) paired Near-Infrared (NIR) image for each frame, iii.) larger label set with a 4-level label hierarchy to capture unstructured traffic conditions. We benchmark state-of-the-art models for semantic segmentation in IDD-AW. We also propose a new metric called “Safe mean Intersection over Union (Safe mIoU)” for hierarchical datasets which penalizes dangerous mispredictions that are not captured in the traditional definition of mean Intersection over Union (mIoU). The results show that IDD-AW is one of the most challenging datasets to date for these tasks