Research

Publications

Research from CHARM’s faculty, students, and collaborators across human‑centered AI, interaction design, and machine learning.

2026

  1. Bias at the End of the Score

    Salma Abdel Magid, Grace Guo, Esin Tureci, Amaya Dharmasiri, Vikram V. Ramaswamy, Hanspeter Pfister, Olga Russakovsky

    IEEE CVPR 2026

  2. Inverse Transition Learning: Learning Dynamics from Demonstrations

    Leo Benac, Abhishek Sharma, Sonali Parbhoo, Finale Doshi-Velez

    AISTATS 2026

  3. Vidmento: Creating Video Stories through Context-Aware Expansion with Generative Video

    Catherine Yeh, Anh Truong, Mira Dontcheva, Bryan Wang

    CHI 2026

  4. MnemoMaker: Creator, Curator, or Something Else? Exploring Human-AI Mnemonic Co-Creation

    Olivia Seow, Elena Sajno, Dongho Shin, Pattie Maes, Samantha W. T. Chan

    CHI 2026 Extended Abstracts (Interactive Demo)

  5. How Notations Evolve: A Historical Analysis with Implications for Supporting User-Defined Abstractions

    Jingyue Zhang, J.D. Zamfirescu-Pereira, Elena L. Glassman, Damien Masson, Ian Arawjo

    CHI 2026

  6. A Paradigm for Creative Ownership

    Tejaswi Polimetla, Katy Ilonka Gero, Elena L. Glassman

    CHI 2026

  7. Meta-HCI: Practising Reflection in HCI Research

    Annika Kaltenhauser, James Peter Arnéra, Amelie Unger, Sophia Ppali, Niels van Berkel, Benjamin Tag, Elena L. Glassman, Phoebe Sengers, Simo Hosio, Jonas Oppenlaender

    CHI 2026 Extended Abstracts (Meetup)

  8. Science and Technology for Augmenting Reading (STAR)

    Tal August, Andrew Head, Alexa Siu, Elena L. Glassman, Jonathan K. Kummerfeld, Joseph Chee Chang, Lucy Lu Wang, Marti A. Hearst

    CHI 2026 Extended Abstracts (Workshop)

  9. Overview of the BRIDGE system

    BRIDGE: Borderless Reconfiguration for Inclusive and Diverse Gameplay Experience via Embodiment Transformation

    Hayato Saiki, Chunggi Lee, Hikari Takahashi, Tica Lin, Hidetada Kishi, Kaori Tachibana, Yasuhiro Suzuki, Hanspeter Pfister, Kenji Suzuki

    CHI 2026 Best Paper Award

    Training resources for parasports are limited, reducing opportunities for athletes and coaches to engage with sport-specific movements and tactical coordination. To address this gap, we developed BRIDGE, a system that integrates a reconstruction pipeline, which detects and tracks players from broadcast video to generate 3D play sequences, with an embodiment-aware visualization framework that decomposes head, trunk, and wheelchair base orientations to represent attention, intent, and mobility. We evaluated BRIDGE in two controlled studies with 20 participants (10 national wheelchair basketball team players and 10 amateur players).The results showed that BRIDGE significantly enhanced the perceived naturalness of player postures and made tactical intentions easier to understand. In addition, it supported functional classification by realistically conveying players’ capabilities, which in turn improved participants’ sense of self-efficacy. This work advances inclusive sports learning and accessible coaching practices, contributing to more equitable access to tactical resources in parasports.

  10. Funding AI for Good paper thumbnail

    Funding AI for Good: A Call for Meaningful Engagement

    Lin, Hongjin; Kawakami, Anna; D’Ignazio, Catherine; Holstein, Kenneth; Gajos, Krzysztof Z

    CHI 2026

    Artificial Intelligence for Social Good (AI4SG) is a growing area that explores AI’s potential to address social issues, such as public health. Yet prior work has shown limited evidence of its tangible benefits for intended communities, and projects frequently face real‑world deployment and sustainability challenges. We conducted a reflexive thematic analysis of 35 funding documents, representing about $410 million USD in total investments.

  11. Beyond Anthropomorphism: a Spectrum of Interface Metaphors for LLMs

    Jianna So, Connie Cheng, Sonia Krishna Murthy

    CHI 2026

  12. Nonvisual Support for Understanding and Reasoning about Data Structures

    Brianna L. Wimer, Ritesh Kanchi, Kaija Frierson, Venkatesh Potluri, Ronald Metoyer, Jennifer Mankoff, Miya Natsuhara, Matt X. Wang

    CHI 2026

  13. “It just requires so much more creativity”: Barriers and Workarounds to Gathering Information for AI Contestation

    Sohini Upadhyay, Dasha Pruss, Alicia DeVrio, Krzysztof Z. Gajos, Naveena Karusala

    CHI 2026

  14. Novel Web-Based Technology to Promote Goal-Setting in Complex Chronic Illness: Randomized Controlled Trial

    Lin, Jody; Huber, Bernd; Amir, Ofra; Assis-Hassid, Shiri; Gehrmann, Sebastian; Gajos, Krzysztof; Grosz, Barbara; Sanders, Lee

    JMIR Hum Factors, vol. 13, pp. e70402, 2026

  15. ViSTAR: Virtual Skill Training with Augmented Reality with 3D Avatars and LLM coaching agent.

    Chunggi Lee, Hayato Saiki, Tica Lin, Eiji Ikeda, Kenji Suzuki, Chen Zhu-Tian, Hanspeter Pfister

    CHI 2026

  16. Federated ADMM from Bayesian Duality

    Thomas Möllenhoff, Siddharth Swaroop, Finale Doshi-Velez, Mohammad Emtiyaz Khan

    ICLR 2026

  17. Virtual Multiplex Staining for Histological Images Using a Marker-wise Conditioned Diffusion Model

    Hyun-Jic Oh, Junsik Kim, Zhiyi Shi, Yichen Wu, Yu-An Chen, Peter K. Sorger, Hanspeter Pfister, and Won-Ki Jeong

    AAAI 2026

  18. 2025

    1. LangSplatV2: High-dimensional 3D Language Gaussian Splatting with 450+ FPS

      Wanhua Li, Yujie Zhao, Minghan Qin, Liu Y, Cai Y, Chuang Gan, Hanspeter Pfister

      NeurIPS 2025

    2. On the Effective Horizon of Inverse Reinforcement Learning

      Yiqing Xu, Finale Doshi-Velez, David Hsu

      AAMAS 2025

    3. VAIR: Visual Analytics for Injury Risk Exploration in Sports

      Chunggi Lee, Ut Gong, Tica Lin, Stefanie Zollmann, Scott A Epsley, Adam Petway, Hanspeter Pfister

      IEEE VIS 16th Workshop on Visual Analytics in Healthcare (VAHC), 2025.

    4. Estimating Upper Extremity Fugl-Meyer Assessment Scores From Reaching Motions Using Wearable Sensors

      Yu Meng Zhou, Nihal Raman, Tommaso Proietti, James Arnold, Prabhat Pathak, David Pont-Esteban, Kristin Nuckolsand, Kelly Rishe, Finale Doshi-Velez, David Lin, Conor Walsh

      IEEE 2025

    5. MoMo – Combining Neuron Morphology and Connectivity for Interactive Motif Analysis in Connectomes

      Shewarega MF, Troidl J, Rodriguez OA, Dindoost M, Harth P, Haberkern H, Stegmaier J, Bader D, Pfister H

      IEEE VIS 2025

    6. SynAnno: Interactive Guided Proofreading of Synaptic Annotations

      Leander Lauenburg, Jakob Troidl, Adam Gohain, Zudi Lin, Hanspeter Pfister, Donglai Wei

      IEEE VIS 2025

    7. Creative Writers’ Attitudes on Writing as Training Data for Large Language Models thumbnail

      Creative Writers’ Attitudes on Writing as Training Data for Large Language Models

      Katy Ilonka Gero, Meera Desai, Carly Schnitzler, Nayun Eom, Jack Cushman, Elena L. Glassman

      CHI 2025 Best Paper Award

      The use of creative writing as training data for large language models (LLMs) is highly contentious and many writers have expressed outrage at the use of their work without consent or compensation. In this paper, we seek to understand how creative writers reason about the real or hypothetical use of their writing as training data. We interviewed 33 writers with variation across genre, method of publishing, degree of professionalization, and attitudes toward and engagement with LLMs. We report on core principles that writers express (support of the creative chain, respect for writers and writing, and the human element of creativity) and how these principles can be at odds with their realistic expectations of the world (a lack of control, industry-scale impacts, and interpretation of scale). Collectively these findings demonstrate that writers have a nuanced understanding of LLMs and are more concerned with power imbalances than the technology itself.

    8. Supporting Co-Adaptive Machine Teaching through Human Concept Learning and Cognitive Theories thumbnail

      Supporting Co-Adaptive Machine Teaching through Human Concept Learning and Cognitive Theories

      Simret Araya Gebreegziabher, Yukun Yang, Elena L. Glassman, Toby Jia-Jun Li

      CHI 2025 Best Paper Award

      An important challenge in interactive machine learning, particularly in subjective or ambiguous domains, is fostering bi-directional alignment between humans and models. Users teach models their concept definition through data labeling, while refining their own understandings throughout the process. To facilitate this, we introduce Mocha, an interactive machine learning tool informed by two theories of human concept learning and cognition. First, it utilizes a neuro-symbolic pipeline to support Variation Theory based counterfactual data generation. By asking users to annotate counterexamples that are syntactically and semantically similar to already-annotated data but predicted to have different labels, the system can learn more effectively while helping users understand the model and reflect on their own label definitions. Second, Mocha uses Structural Alignment Theory to present groups of counterexamples, helping users comprehend alignable differences between data items and annotate them in batch. We validated Mocha’s effectiveness and usability through a lab study with 18 participants.

    9. SEAL: Spatially-resolved Embedding Analysis with Linked Imaging Data

      Simon Warchol, Grace Guo, Johannes Knittel, Dan Freeman, Usha Bhalla, Jeremy L Muhlich, Peter K. Sorger, Hanspeter Pfister

      IEEE VIS 2025

    10. niiv: Interactive Self-supervised Neural Implicit Isotropic Volume Reconstruction

      Jakob Troidl, Yiqing Liang, Johanna Beyer, Mojtaba Tavakoli, Johann Danzl, Markus Hadwiger, Hanspeter Pfister, James Tompkin

      MICCAI Workshop on Efficient Medical AI (EMA), 2025

    11. To Recommend or Not to Recommend: Designing and Evaluating AI-Enabled Decision Support for Time-Critical Medical Events

      Mastrianni, Angela; Kim, Mary Suhyun; Sullivan, Travis M.; Sippel, Genevieve Jayne; Burd, Randall S.; Gajos, Krzysztof Z.; Sarcevic, Aleksandra

      Proc. ACM Hum.-Comput. Interact, vol. 9, iss. CSCW2, 2025.

    12. Transparent Trade-offs between Properties of Explanations

      Hiwot Belay Tadesse, Alihan Hüyük, Yaniv Yacoby, Weiwei Pan, Finale Doshi-Velez

      UAI 2025

    13. Frenet-Serret Frame-based Decomposition for Part Segmentation of 3D Curvilinear Structures

      Leslie Gu, Jason Ken Adhinarta, Mikhail Bessmeltsev, Jiancheng Yang, Yongjie Jessica Zhang, Wenjie Yin, Daniel Berger, Jeff Lichtman, Hanspeter Pfister, Donglai Wei

      IEEE Transactions on Medical Imaging, 2025

    14. The State of Single-Cell Atlas Data Visualization in the Biological Literature

      Mark S Keller, Eric Mörth, Thomas C Smits, Simon Warchol, Grace Guo, Qianwen Wang, Robert Krueger, Hanspeter Pfister, Nils Gehlenborg

      IEEE Computer Graphics and Applications, 2025

    15. Personalising AI assistance based on overreliance rate in AI-assisted decision making

      Siddharth Swaroop, Zana Buçinca, Krzysztof Z. Gajos, Finale Doshi-Velez

      IUI 2025

    16. A connectomic resource for neural cataloguing and circuit dissection of the larval zebrafish brain

      Mariela D. Petkova, Michał Januszewski, Tim Blakely, Kristian J. Herrera, Gregor F.P. Schuhknecht, Robert Tiller, Jinhan Choi, Richard L. Schalek, Jonathan Boulanger-Weill, Adi Peleg, Yuelong Wu, Shuohong Wang, Jakob Troidl, Sumit Kumar Vohra, Donglai Wei, Zudi Lin, Armin Bahl, Juan Carlos Tapia, Nirmala Iyer, Zachary T. Miller, Kathryn B. Hebert, Elisa C. Pavarino, Milo Taylor, Zixuan Deng, Moritz Stingl, Dana Hockling, Alina Hebling, Ruohong C. Wang, Lauren L. Zhang, Sam Dvorak, Zainab Faik, Kareem I. King Jr., Pallavi Goel, Julian Wagner-Carena, David Aley, Selimzhan Chalyshkan, Dominick Contreas, Xiong Li, Akila V. Muthukumar, Marina S. Vernaglia, Teodoro Tapia Carrasco, Sofia Melnychuck, TingTing Yan, Ananya Dalal, James M. DiMartino, Sam Brown, Nana Safo-Mensa, Ethan Greenberg, Michael Cook, Samantha Finley-May, Miriam A. Flynn, Gary Patrick Hopkins, Julie Kovalyak, Meghan Leonard, Alanna Lohff, Christopher Ordish, Ashley L. Scott, Satoko Takemura, Claire Walsh, John J. Walsh, Daniel R. Berger, Hanspeter Pfister, Stuart Berg, Christopher Knecht, Geoffrey W. Meissner, Wyatt Korff, Misha B. Ahrens, Viren Jain, Jeff W. Lichtman, Florian Engert

      bioRxiv, 2025

    17. 4D LangSplat: 4D Language Gaussian Splatting via Multimodal Large Language Models

      Wanhua Li, Renping Zhou, Jiawei Zhou, Y. Song, J. Herter, M. Qin, Gao Huang, Hanspeter Pfister

      CVPR 2025

    18. Decision-Point Guided Safe Policy Improvement

      Abhishek Sharma, Leo Benac, Sonali Parbhoo, Finale Doshi-Velez

      AISTATS 2025

    19. Toward Accounting for the Effects of Gender Socialization in Quantitative Research in Human-Computer Interaction

      Hagen, Nazeli; Miratrix, Luke W.; Gajos, Krzysztof Z.

      Interacting with Computers, 2025.

    20. TriSAM: Tri-Plane SAM for zero-shot cortical blood vessel segmentation in VEM images

      Jia Wan, Wanhua Li, Jason Ken Adhinarta, Atmadeep Banerjee, Evelina Sjostedt, Jingpeng Wu, Jeff Lichtman, Hanspeter Pfister, Donglai Wei

      IEEE Journal of Biomedical and Health Informatics, 2025

    21. CTRL-GS: Cascaded Temporal Residue Learning for 4D Gaussian Splatting

      Karly Hou, Wanhua Li, Hanspeter Pfister

      4D Vision Workshop @ CVPR 2025

    22. AbstractExplorer: Leveraging Structure-Mapping Theory to Enhance Comparative Close Reading at Scale

      Ziwei Gu, Joyce Zhou, Nina Lei, Jonathan Kummerfeld, Mahmood Jasim, Narges Mahyar, Elena L. Glassman

      UIST 2025

    23. Integrated Gradients Provides Faithful Language Model Attributions for In-Context Learning

      Theo Datta, Erik Wang, Kayla Huang, Finale Doshi-Velez

      ICLR 2025 Workshop Building Trust

    24. CAVE: Connectome Annotation Versioning Engine

      Sven Dorkenwald, Casey M. Schneider-Mizell, Derrick Brittain, Akhilesh Halageri, Chris Jordan, Nico Kemnitz, Manual A. Castro, William Silversmith, Jeremy Maitin-Shephard, Jakob Troidl, Hanspeter Pfister, Valentin Gillet, Daniel Xenes, J. Alexander Bae, Agnes L. Bodor, JoAnn Buchanan, Daniel J. Bumbarger, Leila Elabbady, Zhen Jia, Daniel Kapner, Sam Kinn, Sam Kinn, Kisuk Lee, Kai Li, Ran Lu, Thomas Macrina, Gayathri Mahalingam, Eric Mitchell, Shanka Subhra Mondal, Shang Mu, Barak Nehoran, Sergiy Popovych, Marc Takeno, Russel Torres, Nicholas L. Turner, William Wong, Jingpeng Wu, Wenjing Yin, Szi-chieh Yu, R. Clay Reid, Nuno Maçarico da Costa, H. Sebastian Seung, Forrest Collman

      Nature Methods, 2025

    25. Global Neuron Shape Reasoning with Point Affinity Transformers

      Jakob Troidl, Johannes Knittel, Wanhua Li, Fangneng Zhan, Hanspeter Pfister, Srinivas Turaga

      bioRxiv, 2025

    26. Semantic Commit: Helping Users Update Intent Specifications for AI Memory at Scale

      Priyan Vaithilingam, Munyeong Kim, Frida-Cecilia Acosta-Parenteau, Daniel Lee, Amine Mhedhbi, Elena L Glassman, Ian Arawjo

      UIST 2025

    27. Addressing persistent challenges in digital image analysis of cancer tissue: resources developed from a hackathon

      Sandhya Prabhakaran, Clarence Yapp, Gregory J Baker, Johanna Beyer, Young Hwan Chang, Allison L Creason, Robert Krueger, Jeremy Muhlich, Nathan Heath Patterson, Kevin Sidak, Damir Sudar, Adam J Taylor, Luke Ternes, Jakob Troidl, Xie Yubin, Artem Sokolov, Darren R Tyson

      Molecular Oncology, 2025

    28. Contrastive Explanations That Anticipate Human Misconceptions Can Improve Human Decision-Making Skills

      Zana Buçinca, Siddharth Swaroop, Amanda E. Paluch, Krzysztof Z. Gajos, Finale Doshi-Velez

      CHI 2025

    29. SportsBuddy: Designing and Evaluating an AI-Powered Sports Video Storytelling Tool Through Real-World Deployment

      Tica Lin, Ruxun Xiang, Gardenia Liu, Divyanshu Tiwari, Meng-Chia Chiang, Chenjiayi Ye, Hanspeter Pfister, Chen Zhu-Tian

      IEEE PacificVis 2025

    30. Extending reinforcement Learning-Driven Personalized Health Interventions to Multiple Health Behavioral Change Goals

      Samantha Marks, Michelle Chang, Weiwei Pan, Susan Murphy, Finale Doshi-Velez

      MOSS Workshop @ ICML 2025

    31. SD-LoRA: Scalable Decoupled Low-Rank Adaptation for Class Incremental Learning

      Yichen Wu, H.M. Piao, L.K. Huang, R.Z. Wang, Wanhua Li, Hanspeter Pfister, D.Y. Meng, K.D. Ma, Y. Wei

      ICLR 2025

    32. Understanding the Relationship between Prompts and Response Uncertainty in Large Language Models

      Ze Yu Zhang, Arun Verma, Finale Doshi-Velez, Bryan Kian Hsiang Low

      ICLR 2025 Workshop: Quantify Uncertainty and Hallucination in Foundation Models: The Next Frontier in Reliable AI

    33. Tree of Attributes Prompt Learning for Vision-Language Models

      Tong Ding, Wanhua Li, Zhongqi Miao, Hanspeter Pfister

      ICLR 2025

    34. Unleashing the Power of Task-Specific Directions in Parameter Efficient Fine-tuning

      Chongjie Si, Zhiyi Shi, Shifan Zhang, Xiaokang Yang, Hanspeter Pfister, Wei Shen

      ICLR 2025

    35. Connecting Federated ADMM to Bayes

      Siddharth Swaroop, Mohammad Emtiyaz Khan, Finale Doshi-Velez

      ICLR 2025

    36. Connecting Federated ADMM to Bayes

      Siddharth Swaroop, Mohammad Emtiyaz Khan, Finale Doshi-Velez

      ICLR 2025

    37. Law is vulnerable to AI influence; interface design can help

      Aileen Nielsen, Chelse Swoopes, Elena L. Glassman

      SSRN Preprint 2025

    38. Bridging Ontologies of Neurological Conditions: Towards Patient-centered Data Practices in Digital Phenotyping Research and Design thumbnail

      Bridging Ontologies of Neurological Conditions: Towards Patient-centered Data Practices in Digital Phenotyping Research and Design

      Jianna So, Faye Yang, Krzysztof Z. Gajos, Naveena Karusala, Anoopum S. Gupta

      Proceedings of the ACM on Human-Computer Interaction, Honorable Mention

      Amidst the increasing datafication of healthcare, deep digital phenotyping is being explored in clinical research to gather comprehensive data that can improve understanding of neurological conditions. However, participants currently do not have access to this data due to researchers’ apprehension around whether such data is interpretable or useful. This study focuses on patient perspectives on the potential of deep digital phenotyping data to benefit people with neurodegenerative diseases, such as ataxias, Parkinson’s disease, and multiple system atrophy. We present an interview study (n=12) to understand how people with these conditions currently track their symptoms and how they envision interacting with their deep digital phenotyping data. We describe how participants envision the utility of this deep digital phenotyping data in relation to multiple stages of disease and stakeholders, especially its potential to bridge different and sometimes conflicting understandings of their condition. Looking towards a future in which patients have increased agency over their data and can use it to inform their care, we contribute implications for shaping patient-driven clinical research practices and deep digital phenotyping tools that serve a multiplicity of patient needs.

    39. Leveraging Variation Theory in Counterfactual Data Augmentation for Optimized Active Learning thumbnail

      Leveraging Variation Theory in Counterfactual Data Augmentation for Optimized Active Learning

      Simret Araya Gebreegziabher, Kuangshi Ai, Zheng Zhang, Elena L. Glassman*, Toby Jia-Jun Li*

      ACL 2025

      Active Learning (AL) allows models to learn interactively from user feedback. However, only annotating existing samples may hardly benef it the model’s generalization. Moreover, AL commonly faces a cold start problem due to insufficient annotated data for effective sample selection. To address this, we introduce a counterfactual data augmentation approach inspired by Variation Theory, a theory of human concept learning that emphasizes the essential features of a concept by focusing on what stays the same and what changes. We use a neuro-symbolic pipeline to pinpoint key conceptual dimensions and use a large language model (LLM) to generate targeted variations along those dimensions. Through a text classification experiment, we show that our approach achieves significantly higher performance when there are fewer annotated data, showing its capability to address the cold start problem in AL. We also find that as the annotated training data gets larger, the impact of the generated data starts to diminish. This work demonstrates the value of incorporating human learning theories into the design and optimization of AL.

    40. Designing a Dashboard for Transparency and Control of Conversational AI thumbnail

      Designing a Dashboard for Transparency and Control of Conversational AI

      Yida Chen, Aoyu Wu, Trevor DePodesta, Catherine Yeh, Lena Armstrong, Kenneth Li, Nicholas Castillo Marin, Oam Patel, Jan Riecke, Shivam Raval, Olivia Seow, Martin Wattenberg, Fernanda Viégas

      ICML 2025

      Conversational LLMs function as black box systems, leaving users guessing about why they see the output they do. This lack of transparency is potentially problematic, especially given concerns around bias and truthfulness. To address this issue, we present an end-to-end prototype—connecting interpretability techniques with user experience design—that seeks to make chatbots more transparent. We begin by showing evidence that a prominent open-source LLM has a “user model”: examining the internal state of the system, we can extract data related to a user’s age, gender, educational level, and socioeconomic status. Next, we describe the design of a dashboard that accompanies the chatbot interface, displaying this user model in real time. The dashboard can also be used to control the user model and the system’s behavior. Finally, we discuss a study in which users conversed with the instrumented system. Our results suggest that users appreciate seeing internal states, which helped them expose biased behavior and increased their sense of control. Participants also made valuable suggestions that point to future directions for both design and machine learning research. The project page and video demo of our TalkTuner system are available at bit.ly/talktuner-project-page .

    41. Counterfactual Explanations May Not Be the Best Algorithmic Recourse Approach thumbnail

      Counterfactual Explanations May Not Be the Best Algorithmic Recourse Approach

      Sohini Upadhyay, Himabindu Lakkaraju, Krzysztof Z. Gajos

      IUI 2025

      Algorithmic recourse is a rapidly developing subfield in explainable AI (XAI) concerned with providing individuals subject to adverse high-stakes algorithmic outcomes with explanations indicating how to reverse said outcomes. While XAI research in the machine learning community doesn’t confine itself to counterfactual explanations, its algorithmic recourse subfield does, adopting the assumption that the optimal way to provide recourse is through counterfactual explanations. Though there has been extensive human-AI interaction research on explanations, translating these findings to the algorithmic recourse setting is non-obvious due to meaningful problem setting differences, leaving the question of whether counterfactuals are the most optimal explanation paradigm for recourse unanswered. While intuitively satisfying, the prescriptive nature of counterfactuals makes them vulnerable to poor outcomes when circumstances unknown to the decision-making and explanation generating algorithms affect re-application strategies. With these concerns in mind, we designed a series of experiments comparing different explanation methods in the recourse setting, explicitly incorporating scenarios where circumstances unknown to the decision-making and explanation algorithms affect re-application strategies. In Experiment 1, we compared counterfactuals with reason codes, a simple feature-based explanation, finding that they both yield comparable re-application success, and that reason codes led to better user outcomes when unknown circumstances had a high impact on re-application strategies. In Experiment 2, we sought to improve on reason code outcomes, comparing them to feature attributions, a more informative feature-based explanation, but found no improvements. Finally, in Experiment 3, we aimed to improve on reason code outcomes with a multiple counterfactual explanation condition, finding that multiple counterfactuals led to higher re-application success but still resulted in comparatively worse user outcomes in the face of high impact unknown circumstances. Taken together, these findings call into question whether the standard counterfactual paradigm is the best approach for the algorithmic recourse problem setting.

    42. Bayesian Hypothesis Testing Policy Regularization thumbnail

      Bayesian Hypothesis Testing Policy Regularization

      Sarah Rathnam, Finale Doshi-Velez, Susan Murphy

      RLC 2025, ICML 2025

      In reinforcement learning (RL), sparse feedback makes it difficult to target long-term outcomes, often resulting in high-variance policies. Real-world interventions instead rely on prior study data, expert input, or short-term proxies to guide exploration. In this work, we propose Bayesian Hypothesis Testing Policy Regularization (BHTPR), a method that integrates a previously-learned policy with a policy learned online to speed up learning in such settings. BHTPR applies the inductive bias that the prior study data matches the current study environment in some states but is incorrect in others. We use Bayesian hypothesis testing to determine, state by state, when to transfer the prior policy and when to rely on online learning.

    43. CorpusStudio paper thumbnail

      CorpusStudio: Surfacing Emergent Patterns In A Corpus Of Prior Work While Writing

      Hai Dang, Chelse Swoopes, Daniel Buschek, Elena L. Glassman

      CHI ’25 · April 26–May 01, 2025 · Yokohama, Japan

      Many communities, including the scientific community, develop implicit writing norms. Understanding them is crucial for effective communication with that community. Writers gradually develop an implicit understanding of norms by reading papers and receiving feedback on their writing. However, it is difficult to both externalize this knowledge and apply it to one’s own writing. We propose two new writing support concepts that reify document and sentence‑level patterns in a given text corpus.

    44. ChatGPT paper thumbnail

      “ChatGPT, Don’t Tell Me What to Do”: Designing AI for Context Analysis in Humanitarian Frontline Negotiations

      Zilin Ma, Yiyang Mei, Claude Bruderlein, Krzysztof Z. Gajos, Weiwei Pan

      CHIWORK · ACM Press · Forthcoming

      Frontline humanitarian negotiators are increasingly exploring ways to use AI tools in their workflows. However, current AI tools in negotiation primarily focus on outcomes, neglecting crucial aspects of the negotiation process. Through iterative user‑centric design with experienced frontline negotiators (n=32), we found that flexible tools that enable contextualizing cases and exploring options are more effective than those providing direct recommendations.

    45. Optimizing Explanations paper thumbnail

      Optimizing Explanations: Nuances Matter When Evaluation Metrics Become Loss Functions

      Jonas B. Raedler, Hiwot Belay Tadesse, Weiwei Pan, Finale Doshi‑Velez

      MOSS Workshop @ ICML 2025

      Recent work has introduced a framework that allows users to directly optimize explanations for desired properties and their trade‑offs. While powerful in principle, this method repurposes evaluation metrics as loss functions. We study how different robustness metrics influence the outcome of explanation optimization, and find that the choice of metric can lead to highly divergent explanations, particularly in higher‑dimensional settings.