Rafal Kocielnik

I am a Postdoctoral Researcher at Caltech where I am mentored by Anima Anandkumar, Mike Alvarez, and Andrew Hung. I received my PhD in Human Centered Design & Engineering from the University of Washington, where I was advised by Gary Hsieh.

My primary research interests:

  • Human-Centered AI – Designing interactions and building AI systems that promote social good, particularly in the contexts of social media, gaming, and health.
  • Human-AI Alignment – Exploring how generative AI can be aligned with human values by leveraging insights from social science and human-centered design.
  • AI Transparency & User Empowerment – Creating tools that enable both everyday users and domain experts to understand the capabilities and limitations of AI systems.

Over the summers, I've been lucky to intern with Jonathan Bragg and Doug Downey at Allen Institute for AI, Saleema Amershi and Andrés Monroy-Hernández at Microsoft Research Redmond, and Daniel Avrahami at Fuji-Xerox Palo Alto Research Lab.

rafalko [at] caltech.edu

News

Jul 2025 "Tracing Human-like Traits in LLMs" accepted at ICML workshop on Models of Human Feedback for AI Alignment
Jun 2025 Two papers in collaboration with Activision accepted for publication:
"Causal Estimates of Effective Moderation in Competitive Action Games" as full paper at CHI Play 2025
"Bandit Algorithms for Toxicity Detection in Games" at IEEE Access
Mar 2025 Seminar talk: "Bridging AI Innovation & Human-Centered Design," at CST, Cambridge University, UK.
Feb 2025 Congrats to my mentee Arushi on the CS PhD offers from Stanford and Harvard!
Jan 2025 Seminar talk: "Human-Centered AI: From Clinical Education Support to Enabling End-user Inspection of AI," at CS Department, Barnard College, Columbia University.
Dec 2024 "Human AI Collaboration for Unsupervised Categorization of Live Surgical Feedback" published at Nature npj | Digital Medicine
Nov 2024 Two proceedings papers at ML4H 2024:
"Multi-Modal Self-Supervised Learning for Surgical Feedback Effectiveness Assessment" Best Paper Award
"Automating Feedback Analysis in Surgical Training: Detection, Categorization, and Assessment"
Oct 2024 "ChatGPT Based Data Augmentation for Improved Parameter-Efficient Debiasing of LLMs" published at COLM 2024

Selected Publications

Tracing Human-like Traits in LLMs: Origins, Real-World Manifestation, and Controllability
Rafal Kocielnik*, Pengrui Han*, Peiyang Song, Ramit Debnath, Dean Mobbs, Animashree Anandkumar, R. Michael Alvarez
ICML Workshop - Models of Human Feedback for AI Alignment , 2025
OpenReview

Online Moderation in Competitive Action Games: How Intervention Affects Player Behaviors
Rafal Kocielnik*, Zhuofang Li*, Mitchell Linegar, Deshawn Sambrano, Fereshteh Soltani, Min Kim, Nabiha Naqvie, Grant Cahill, Animashree Anandkumar, R. Michael Alvarez
CHI Play, 2025 (in print)
arXiv

Human AI Collaboration for Unsupervised Categorization of Live Surgical Feedback
Rafal Kocielnik, Cherine H. Yang, Runzhuo Ma, Steven Y. Cen, Elyssa Y. Wong, Timothy N. Chu, J. Everett Knudsen, Peter Wager, John Heard, Umar Ghaffar, Animashree Anandkumar, Andrew J. Hung
Nature npj | Digital Medicine, 2024
Publisher

Artificial Intelligence-Based Video Feedback to Improve Novice Performance on Robotic Suturing Skills: A Pilot Study
Runzhuo Ma, Dani Kiyasseh, Jasper A Laca, Rafal Kocielnik, Elyssa Y Wong, Timothy N Chu, Steven Cen, Cherine H Yang, Istabraq S Dalieh, Taseen F Haque, Mitch G Goldenberg, Xiuzhen Huang, Anima Anandkumar, Andrew J Hung
Journal of Endourology, 2024
Publisher

Deep Multimodal Fusion for Surgical Feedback Classification
Rafal Kocielnik, Elyssa Y. Wong, Timothy N. Chu, Lydia Lin, De-An Huang, Jiayun Wang, Anima Anandkumar, Andrew J. Hung
Machine Learning for Health (ML4H), 2023 Best Proceedings Paper
Publisher

Exploring Social Bias in Downstream Applications of Text-to-Image Foundation Models
Adhithya Saravanan, Rafal Kocielnik, Roy Jiang, Pengrui Han, Anima Anandkumar
NeurIPS workshop on Failure Modes in the Age of Foundation Models, 2023
(accepted for special issue of PMLR)
arXiv OpenReview

BiasTestGPT: Using ChatGPT for Social Bias Testing of Language Models
Rafal Kocielnik, Shrimai Prabhumoye, Vivian Zhang, Roy Jiang, R. Alvarez Michael, Anima Anandkumar
ICML workshop on Deployment Challenges for Generative AI 2023
arXiv HuggingFace Tool Project Website Dataset

Large Language Models and Political Science
Mitchell Linegar, Rafal Kocielnik, R Michael Alvarez
Frontiers in Political Science, 2023
Publisher

Can You Label Less by Using Out-of-Domain Data? Active & Transfer Learning with Few-shot Instructions
Rafal Kocielnik*, Sara Kangaslahti*, Shrimai Prabhumoye, Meena Hari, Michael Alvarez, Anima Anandkumar
NeurIPS 2023 Transfer Learning for Natural Language Processing Workshop
Publisher Medium Blog

From Who You Know to What You Read: Augmenting Scientific Recommendations with Implicit Social Networks
Hyeonsu B Kang, Rafal Kocielnik, Andrew Head, Jiangjiang Yang, Matt Latzke, Aniket Kittur, Daniel S Weld, Doug Downey, Jonathan Bragg
CHI 2022
Publisher

Using real-time feedback to improve surgical performance on a robotic tissue dissection task
Jasper A Laca, Rafal Kocielnik, Jessica H Nguyen, Jonathan You, Ryan Tsang, Elyssa Y Wong, Andrew Shtulman, Anima Anandkumar, Andrew J Hung
European Urology Open Science, 2022
Publisher

Can I Talk to You about Your Social Needs? Understanding Preference for Conversational User Interface in Health
Rafal Kocielnik, Raina Langevin, James S George, Shota Akenaga, Amelia Wang, Darwin P Jones, Alexander Argyle, Callan Fockele, Layla Anderson, Dennis T Hsieh, Kabir Yadav, Herbert Duber, Gary Hsieh, Andrea L Hartzler
Conversational User Interfaces, 2021 Best Paper Honorable Mention
Publisher

HarborBot: A Chatbot for Social Needs Screening
Rafal Kocielnik, Elena Agapie, Alexander Argyle, Dennis T Hsieh, Kabir Yadav, Breena Taira, Gary Hsieh
AMIA Annual Symposium Proceedings 2019
Publisher

Reflection Companion: A Conversational System for Engaging Users in Reflection on Physical Activity
Rafal Kocielnik, Lillian Xiao, Daniel Avrahami, Gary Hsieh
UbiComp / IMWUT 2018
High Impact: 240+ citations
Publisher

Designing for Workplace Reflection: A Chat and Voice-Based Conversational Agent
Rafal Kocielnik, Daniel Avrahami, Jennifer Marlow, Di Lu, Gary Hsieh
Design of Interactive Systems 2018
Publisher

Reciprocity and Donation: How Article Topic, Quality and Dwell Time Predict Banner Donation on Wikipedia
Rafal Kocielnik, Os Keyes, Jonathan T Morgan, Dario Taraborelli, David W McDonald, Gary Hsieh
CSCW 2018
Publisher

Calendar. help: Designing a Workflow-Based Scheduling Agent with Humans in the Loop Authors
Justin Cranshaw, Emad Elwany, Todd Newman, Rafal Kocielnik, Bowen Yu, Sandeep Soni, Jaime Teevan, Andrés Monroy-Hernández
CHI 2017
Publisher

You get who you pay for: The impact of incentives on participation bias
Gary Hsieh, Rafal Kocielnik
CSCW 2016 Best Paper Award
Publisher

Smart technologies for long-term stress monitoring at work
Rafal Kocielnik, Natalia Sidorova, Fabrizio Maria Maggi, Martin Ouwerkerk, Joyce HDM Westerink
International Symposium on Computer-based Medical Systems 2013
High Impact: 190+ citations
Publisher