current research (independent researcher) Inference-time preference elicitation in multi-agent LLMs Bringing ideas from social choice to how multi-agent LLM systems reach collective answers. Ongoing work. Improving LLMs through multi-agent fine-tuning Improving large language models by fine-tuning groups of agents on their own interactions. Ongoing work. masters @ Penn State Better preference labels for RLHF Using surprisingly popular voting to label reward-model data when most labelers are wrong. Part of my MS thesis. Surprisingly popular voting Recovering the ground truth from crowds of experts and non-experts. NeurIPS 2024, WWW 2025 and KDD 2026. Breast cancer detection from mammograms Combining CNN and attention-based (DeiT) image models with patient metadata to detect breast cancer in screening mammograms. undergraduate @ NIT Rourkela Human gait as a cognitive state predictor Undergraduate thesis predicting cognitive state from 3D gait captured by eight Kinect sensors. Multi-Kinect camera calibration Pairwise and octagonal calibration methods for eight Kinect v2 sensors. CVIP 2021. Hand gesture password recognition A dynamic hand gesture-based password system that works with an ordinary laptop webcam. ICCIS 2019. Face detection and tracking Real-time face tracking with Viola-Jones detection and Arduino-controlled pan and tilt servos. Face recognition with PCA Eigenface-based face recognition in MATLAB, my first machine learning project.