Affiliated Research

This page showcases an evolving list of affiliated research projects that have utilized the UChicago DSI Cluster, highlighting current work that benefits from the cluster’s computing resources.

Publications

2025

• On the effectiveness and generalization of race representations for debiasing high-stakes decisions
  Nguyen, D., & Tan, C.

• Entailment Progressions: A Robust Approach to Evaluating Reasoning Within Larger Discourse
  Shastry, R., Chiril, P., Charney, J., & Uminsky, D.

• Can AI Weather Models Predict the 2024 Dubai Rainfall, a Regional Grey Swan?
  Sun, Y. Q., Hassanzadeh, P., & Shaw, T. A.

• A Century of Inflation Narratives
  Heddaya, M., Tan, C., Voigt, R., Zeng, Q., & Zentefis, A.

• Deep Learning and Non-Invertible Symmetries in Gauge Theories
  Apte, A.

• Cer-Eval: Certifiable and Cost-Efficient Evaluation Framework for LLMs
  Wang, G., Chen, Z., Li, B., & Xu, H.

• The content, structure, and history of English trait words
  Liu, Y., Charlesworth, T., Koch, A., Luttrell, A., & Jackson, J. C.

• Efficient Distributed Optimization under Heavy-Tailed Noise
  Lee, S. H., Zaheer, M., & Li, T.

• Humans and convolutional neural networks prioritize similar visual features in intuitive physics judgments
  Calabro, R., Bhattacharyya, K., Bainbridge, W., & Leong, Y. C.

• Predicting whole-brain neural dynamics from prefrontal cortex functional near-infrared spectroscopy signal during movie-watching
  Gao, S., Nash, R., Burns, S., & Leong, Y. C.

• A Multimodal Seq2Seq Transformer for Predicting Brain Responses to Naturalistic Stimuli
  He, Q., & Leong, Y. C.

• Automated scoring of the Ambiguous Intentions Hostility Questionnaire using fine-tuned large language models.
  Lyu, Y., Combs, D., Neumann, D., & Leong, Y. C.

• Emotional arousal enhances narrative memories through functional integration of large-scale brain networks
  Park, J. S., Gollapudi, K., Ke, J., Nau, M., Pappas, I., & Leong, Y. C.

• Linear representations of political perspective emerge in large language models
  Kim, J., Evans, J., & Schein, A.

• Unraveling Misinformation Propagation in LLM Reasoning
  Feng, Y., Wang, Y., Cui, S., Faltings, B., Lee, M., & Zhou, J.

• AbsenceBench: Language Models Can’t Tell What’s Missing
  Fu, H. Y., Shrivastava, A., Moore, J., West, P., Tan, C., & Holtzman, A.

• Linearly Decoding Refused Knowledge in Aligned Language Models
  Shrivastava, A., & Holtzman, A.

• Data assimilation with machine learning surrogate models: A case study with FourCastNet
  Adrian, M., Sanz-Alonso, D., & Willett, R.

• Evaluating the Goal-Directedness of Large Language Models
  Everitt, T., Garbacea, C., Bellot, A., Richens, J., Papadatos, H., Campos, S., & Shah, R.

• HyPerAlign: Interpretable Personalized LLM Alignment via Hypothesis Generation
  Garbacea, C., & Tan, C.

• MOFA: Discovering Materials for Carbon Capture with a GenAI-and Simulation-Based Workflow
  Yan, X., Hudson, N., Park, H., Grzenda, D., Pauloski, J. G., Schwarting, M., Harb, H., Foreman, S., Knight, C., Gibbs, T., Chard, K., Chaudhuri, S., Tajkhorshid, E., Foster, I., Moosavi, M., Ward, L., & Huerta, E. A.

• Cartesian Equivariant Representations for Learning and Understanding Molecular Orbitals
  King, D., Grzenda, D., Zhu, R., Hudson, N., Foster, I., Cheng, B., & Gagliardi, L.

• Historical Polarization in Legislative Support for Civil Rights in the United States
  Jackson, J. C., Liu, Y., & Kteily, N. S.

• Prejudice Tied to State Centralization in Historical Societies
  Dillion, D., Liu, Y., Chen, Y., Watts, J., Zhao, C., Baral, S., Bucker, W., Atari, M., Kteily, N., & Jackson, J. C.

• Rising Moralization in Social Media Discourse
  Puryear, C., Brady, W., Jackson, J. C., Leong, Y., & Kteily, N.

• Seeds of Discourse: A Multilingual Corpus of Direct Quotations from African Media on Agricultural Biotechnologies
  Chiril, P., Spreadbury, T., Rock, J. S., Dowd-Uribe, B., & Uminsky, D.

• A Century of Inflation Narratives
  Heddaya, M., Tan, C., Voigt, R., Zeng, Q., & Zentefis, A.

• A machine learning model using clinical notes to identify physician fatigue
  Hsu, C.-C., Obermeyer, Z., & Tan, C.

• Hypobench: Towards systematic and principled benchmarking for hypothesis generation
  Liu, H., Huang, S., Hu, J., Zhou, Y., & Tan, C.

• Relu neural networks with linear layers are biased towards single-and multi-index models
  Parkinson, S., Ongie, G., & Willett, R.

• Prediction-Powered Adaptive Shrinkage Estimation
  Li, S., & Ignatiadis, N.

• Stein’s unbiased risk estimate and Hyvärinen’s score matching
  Ghosh, S., Ignatiadis, N., Koehler, F., & Lee, A.

• Introspective Growth: Automatically Advancing LLM Expertise in Technology Judgment
  Wu, S., Bao, H., Kunievsky, N., & Evans, J. A.

• Language models surface the unwritten code of science and society
  Bao, H., Wu, S., Choi, J., Mao, Y., & Evans, J. A.

• Layers at similar depths generate similar activations across llm architectures
  Wolfram, C., & Schein, A.

• Broad Spectrum Structure Discovery in Large-Scale Higher-Order Networks
  Hood, J., De Bacco, C., & Schein, A.

• Modeling Latent Underdispersion with Discrete Order Statistics
  Lederman, J., & Schein, A.

• Private Retrieval Augmented Generation with Random Projection
  Yao, D., & Li, T.

• Multiple Streams of Relation Extraction: Enriching and Recalling in Transformers
  Nief, T., Reber, D., Richardson, S., & Holtzman, A.

• HypoEval: Hypothesis-Guided Evaluation for Natural Language Generation
  Li, M., Li, H., & Tan, C.

• Multi-frequency progressive refinement for learned inverse scattering
  Melia, O., Tsang, O., Charisopoulos, V., Khoo, Y., Hoskins, J., & Willett, R.

• Hardware Acceleration for HPS Algorithms in Two and Three Dimensions
  Melia, O., Fortunato, D., Hoskins, J., & Willett, R.

• Incorporating Hierarchical Semantics in Sparse Autoencoder Architectures
  Muchane, M., Richardson, S., Park, K., & Veitch, V.

• Concept Incongruence: An Exploration of Time and Death in Role Playing
  Bai, X., Peng, I., Singh, A., & Tan, C.

• Grounded Persuasive Language Generation for Automated Marketing
  Wu, J., Yang, C., Mahns, S., Wang, C., Zhu, H., Fang, F., & Xu, H.

• Tokenized Bandit for LLM Decoding and Alignment
  Shin, S., Yang, C., Xu, H., & Hajiaghayi, M. T.

• How Alignment Shrinks the Generative Horizon
  Yang, C., & Holtzman, A.

• Hierarchical Implicit Neural Emulators
  Jiang, R., Zhang, X., Jakhar, K., Lu, P. Y., Hassanzadeh, P., Maire, M., & Willett, R.

• CLEAR: A Clinically-Grounded Tabular Framework for Radiology Report Evaluation
  Jiang, Y., Chen, C., Wang, S., Li, F., Tang, Z., Mervak, B. M., Chelala, L., Straus, C. M., Chahine, R., Armato, S. G. III, & Tan, C.

• The geometry of categorical and hierarchical concepts in large language models
  Park, K., Choe, Y. J., Jiang, Y., & Veitch, V.

• RATE: Causal Explainability of Reward Models with Imperfect Counterfactuals
  Reber, D., Richardson, S. M., Nief, T., Garbacea, C., & Veitch, V.

• Solving Inverse Problems with Deep Linear Neural Networks: Global Convergence Guarantees for Gradient Descent with Weight Decay
  Laus, H., Parkinson, S., Charisopoulos, V., Krahmer, F., & Willett, R.

• Nonlinear tomographic reconstruction via nonsmooth optimization
  Charisopoulos, V., & Willett, R.

2024

• Causal Micro-Narratives
  Zentefis, A., Tan, C., Heddaya, M., Voigt, R., & Zeng, Q.

• Casesumm: a large-scale dataset for long-context summarization from us supreme court opinions
  Heddaya, M., MacMillan, K., Malani, A., Mei, H., & Tan, C.

• Inconsistency of cross-validation for structure learning in Gaussian graphical models
  Lyu, Z., Tai, W. M., Kolar, M., & Aragam, B.

• Efficient adaptive federated optimization
  Lee, S. H., Sharma, S., Zaheer, M., & Li, T.

• Tilted Sharpness-Aware Minimization
  Li, T., Zhou, T., & Bilmes, J. A.

• Tree Bandits for Generative Bayes
  O’Hagan, S., Kim, J., & Rockova, V.

• Does Editing Provide Evidence for Localization?
  Wang, Z., & Veitch, V.

• Hostile attribution bias shapes neural synchrony in the left ventromedial prefrontal cortex during ambiguous social narratives
  Lyu, Y., Su, Z., Neumann, D., Meidenbauer, K. L., & Leong, Y. C.

• Residual connections harm self-supervised abstract feature learning
  Zhang, X., Jiang, R., Gao, W., Willett, R., & Maire, M.

• Latent intrinsics emerge from training to relight
  Zhang, X., Gao, W., Jain, S., Maire, M., Forsyth, D., & Bhattad, A.

• The linear representation hypothesis and the geometry of large language models
  Park, K., Choe, Y. J., & Veitch, V.

• An Empirical Investigation of Container Building Strategies and Warm Times to Reduce Cold Starts in Scientific Computing Serverless Functions
  Bauer, A., Gonthier, M., Pan, H., Chard, R., Grzenda, D., Straesser, M., Pauloski, J. G., Kamatar, A., Baughman, M., Hudson, N., Foster, I., & Chard, K.

• Sok: On finding common ground in loss landscapes using deep model merging techniques
  Khan, A., Nief, T., Hudson, N., Sakarvadia, M., Grzenda, D., Ajith, A., Pettyjohn, J., Chard, K., & Foster, I.

• Deep Model Merging: The Sister of Neural Network Interpretability–A Survey
  Khan, A., Nief, T., Hudson, N., Sakravadia, M., Grzenda, D., Ajith, A., Pettyjohn, J., Chard, K., & Foster, I.

• BioMAISx: A Corpus for Aspect-Based Sentiment Analysis of Media Representations of Agricultural Biotechnologies in Africa
  Chiril, P., Spreadbury, T., Rock, J. S., Dowd-Uribe, B., & Uminsky, D.

• Quantifying the uniqueness and divisiveness of presidential discourse
  Zhou, K., Meitus, A. A., Chase, M., Wang, G., Mykland, A., Howell, W., & Tan, C.

• Hypothesis generation with large language models
  Zhou, Y., Liu, H., Srivastava, T., Mei, H., & Tan, C.

• The ALCORE Tensor Decomposition for Sparse Count Data
  Hood, J., & Schein, A. J.

• Nested diffusion models using hierarchical latent priors
  Zhang, X., Jiang, R., Willett, R., & Maire, M.

• Deep learning lattice gauge theories
  Apte, A., Córdova, C., Huang, T.-C., & Ashmore, A.

• Literature Meets Data: A Synergistic Approach to Hypothesis Generation
  Liu, H., Zhou, Y., Li, M., Yuan, C., & Tan, C.

• Neural entropy
  Premkumar, A.

• Diffusion Density Estimators
  Premkumar, A.

• Identifying Self-Disclosures of Use, Misuse and Addiction in Community-based Social Media Posts (NAACL 2024 Findings)
  Yang, C., Chakrabarty, T., Hochstatter, K. R., Slavin, M. N., El‑Bassel, N., & Muresan, S.

• When Hindsight is Not 20/20: Testing Limits on Reflective Thinking in Large Language Models (NAACL 2024 Findings)
  Li, Y., Yang, C., & Ettinger, A.

• Training Machine Learning Emulators to Preserve Invariant Measures of Chaotic Attractors
  Lu, P., Jiang, R., Orlova, E., & Willett, R.

• Embed and Emulate: Contrastive representations for simulation-based inference
  Jiang, R., Lu, P. Y., & Willett, R.

• Bonbon alignment for large language models and the sweetness of best-of-n sampling
  Gui, L., Gârbacea, C., & Veitch, V.

• Stabilizing black-box model selection with the inflated argmax.
  Adrian, M., Soloff, J. A., & Willett, R.

• Gpt-4v cannot generate radiology reports yet
  Jiang, Y., Chen, C., Nguyen, D., Mervak, B. M., & Tan, C.

• Beyond reverse KL: Generalizing direct preference optimization with diverse divergence constraints
  Wang, C., Jiang, Y., Yang, C., Liu, H., & Chen, Y.

• On the origins of linear representations in large language models
  Jiang, Y., Rajendran, G., Ravikumar, P. K., Aragam, B., & Veitch, V.

2023

• Measurement in the age of llms: An application to ideological scaling
  O’Hagan, S., & Schein, A.

• Clinical Notes Reveal Physician Fatigue
  Hsu, C.-C., Obermeyer, Z., & Tan, C.

• Training neural operators to preserve invariant measures of chaotic attractors
  Jiang, R., Lu, P. Y., Orlova, E., & Willett, R.

• Rotation-invariant random features provide a strong baseline for machine learning on 3D point clouds
  Melia, O., Jonas, E., & Willett, R.

• Can You Follow Me? Testing Situational Understanding in ChatGPT (EMNLP 2023)
  Yang, C., & Ettinger, A.

• Deep Stochastic Mechanics
  Orlova, E., Ustimenko, A., Jiang, R., Lu, P. Y., & Willett, R.

• Concept algebra for (score-based) text-controlled generative models
  Wang, Z., Gui, L., Negrea, J., & Veitch, V.

• Uncovering meanings of embeddings via partial orthogonality
  Jiang, Y., Aragam, B., & Veitch, V.

• Learning nonparametric latent causal graphs with unknown interventions
  Jiang, Y., & Aragam, B.

2022

• A Unified Causal View of Domain Invariant Representation Learning
  Wang, Z., & Veitch, V.