Shreyas Kadekodi

About

I am a PhD student in Computer Science at UC San Diego working with Berk Ustun. My research focuses on preference aggregation, pluralistic alignment, and selective decision-making: how to represent disagreement in human preferences and when models should abstain.

Previously, I worked with Robin Burke at That Recommender Systems Lab (University of Colorado, Boulder) on fairness in recommender systems. I also worked with Alessandro Roncone at the HIRO Group.

Research

I work on methods for aggregating preferences and producing rankings when human judgments are noisy, incomplete, or disagree. I study how models can represent diverse preferences, expose disagreement, and abstain when there is insufficient agreement.

• Preference aggregation under heterogeneity: partial orders / abstention mechanisms that avoid arbitrary decisions.
• Pluralistic alignment: representing diverse human preferences and preserving meaningful disagreement.
• Selective decision-making: reliable aggregation and prediction with abstention.
• Applications: collective decision-making, recommendation, and human-feedback-based evaluation pipelines.

Featured publication

• ICML 2025

  
  Selective Preference Aggregation
  Shreyas Kadekodi*, Hayden McTavish*, Berk Ustun
  International Conference on Machine Learning (ICML), 2025
  

  Paper PDF Code

  TL;DR: rankings that only reflect what raters actually support.

Public preprints

• LeRoPE: Learnable RoPE Frequencies Improve Language Modeling
  Petros Karypis*, Sean O'Brien*, Shreyas Kadekodi, Rui Zhu, Julian McAuley
  arXiv preprint, 2026
  

  Paper (arXiv)
• Measuring What Matters: Synthetic Benchmarks for Concept Bottleneck Models
  Julian Skirzynski, Harry Cheon, Shreyas Kadekodi, Meredith Stewart, Berk Ustun
  arXiv preprint, 2026
  

  Paper (arXiv)

Earlier peer-reviewed publications

• IROS 2023

  
  CAT-RRT: Motion Planning that Admits Contact One Link at a Time
  Nataliya Nechyporenko, Caleb Escobedo, Shreyas Kadekodi, Alessandro Roncone
  IEEE/RSJ International Conference on Intelligent Robots and Systems, 2023
  

  Paper (arXiv)
• IROS 2022

  
  A Framework for the Systematic Evaluation of Obstacle Avoidance and Object-Aware Controllers
  Caleb Escobedo, Nataliya Nechyporenko, Shreyas Kadekodi, Alessandro Roncone
  IEEE/RSJ International Conference on Intelligent Robots and Systems, 2022
  

  Paper
• CIKM 2021

  
  Librec-auto: A tool for recommender systems experimentation
  Nasim Sonboli, Masoud Mansoury, Ziyue Guo, Shreyas Kadekodi, Weiwen Liu, Zijun Liu, Andrew Schwartz, Robin Burke
  ACM International Conference on Information and Knowledge Management, 2021
  

  Paper (ACM DL)

Contact

Best way to reach me: skadekodi [at] ucsd [dot] edu.