This year, Penn researchers contributed more than 100 papers across the main conference and workshops, covering topics such as large language models, generative AI, reinforcement learning, trustworthy AI, computational biology, neuroscience and applications in healthcare and the physical sciences. Below is the list of Penn-affiliated papers and collaborators presented at NeurIPS 2025.
When Data Can’t Meet: Estimating Correlation Across Privacy BarriersAbhinav Chakraborty, Arnab Auddy, T. Tony Cai
Does Object Binding Naturally Emerge in Large Pretrained Vision Transformers?Yihao Li, Saeed Salehi, Lyle Ungar, Konrad Kording
Mitigating the Privacy–Utility Trade-off in Decentralized Federated Learning via f-Differential PrivacyXiang Li, Chendi Wang, Buxin Su, Qi Long, Weijie J. Su
Optimal Neural Compressors for the Rate-Distortion-Perception TradeoffEric Lei, Hamed Hassani, Shirin Saeedi Bidokhti
ESCA: Contextualizing Embodied Agents via Scene-Graph GenerationJiani Huang, Amish Sethi, Matthew Kuo, Mayank Keoliya, Neelay Velingker, Jungho Jung, Ser-Nam Lim, Ziyang Li, Mayur Naik
STARFlow: Scaling Latent Normalizing Flows for High-resolution Image SynthesisJiatao Gu, Tianrong Chen, David Berthelot, Huangjie Zheng, Yuyang Wang, Ruixiang Zhang, Laurent Dinh, Miguel Ángel Bautista, Joshua M. Susskind, Shuangfei Zhai
On the Empirical Power of Goodness-of-Fit Tests in Watermark DetectionWeiqing He, Xiang Li, Tianqi Shang, Li Shen, Weijie J. Su, Qi Long
Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive FlowsRuixiang Zhang, Shuangfei Zhai, Jiatao Gu, Yizhe Zhang, Huangjie Zheng, Tianrong Chen, Miguel Ángel Bautista, Joshua M. Susskind, Navdeep Jaitly
FORLA: Federated Object-centric Representation Learning with Slot AttentionGuiqiu Liao, Matjaz Jogan, Eric Eaton, Daniel A. Hashimoto
Let Me Think! A Long Chain of Thought Can Be Worth Exponentially Many Short OnesParsa Mirtaheri, Ezra Edelman, Samy Jelassi, Eran Malach, Enric Boix-Adserà
Generalizable, Real-time Neural Decoding with Hybrid State-space ModelsAvery Hee-Woon Ryoo, Nanda H. Krishna, Ximeng Mao, Mehdi Azabou, Eva L. Dyer, Matthew G. Perich, Guillaume Lajoie
Uncertainty-Calibrated Prediction of Randomly Timed Biomarker Trajectories with Conformal BandsVasiliki Tassopoulou, Charis Stamouli, Haochang Shou, George J. Pappas, Christos Davatzikos
Alignment of Large Language Models with Constrained LearningBotong Zhang, Shuo Li, Ignacio Hounie, Osbert Bastani, Dongsheng Ding, Alejandro Ribeiro
Know Thyself by Knowing Others: Learning Neuron Identity from Population ContextVinam Arora, Divyansha Lachi, Ian Jarratt Knight, Mehdi Azabou, Blake Aaron Richards, Cole Lincoln Hurwitz, Josh Siegle, Eva L. Dyer
Conformal Prediction Beyond the Seen: A Missing Mass Perspective for Uncertainty Quantification in Generative ModelsSima Noorani, Shayan Kiyani, George J. Pappas, Hamed Hassani
Composition and Alignment of Diffusion Models Using Constrained LearningShervin Khalafi, Ignacio Hounie, Dongsheng Ding, Alejandro Ribeiro
Real-world Reinforcement Learning of Active Perception BehaviorsEdward S. Hu, Jie Wang, Xingfang Yuan, Fiona Luo, Muyao Li, Gaspard Lambrechts, Oleh Rybkin, Dinesh Jayaraman
Transformers Provably Learn Chain-of-Thought Reasoning with Length GeneralizationYu Huang, Zixin Wen, Aarti Singh, Yuejie Chi, Yuxin Chen
Stochastic Regret Guarantees for Online Zeroth- and First-order Bilevel OptimizationParvin Nazari, Bojian Hou, Davoud Ataee Tarzanagh, Li Shen, George Michailidis
A Scalable, Causal, and Energy-efficient Framework for Neural Decoding with Spiking Neural NetworksGeorgios Mentzelopoulos, Ioannis Asmanis, Konrad Kording, Eva L. Dyer, Kostas Daniilidis, Flavia Vitale
REMI: Reconstructing Episodic Memory During Internally Driven Path PlanningZhaoze Wang, Genela Morris, Dori Derdikman, Pratik Chaudhari, Vijay Balasubramanian
Synthetic-powered Predictive InferenceMeshi Bashari, Roy Maor Lotan, Yonghoon Lee, Edgar Dobriban, Yaniv Romano
Statistical Inference for Gradient Boosting RegressionHaimo Fang, Kevin Tan, Giles Hooker
Imbalances in Neurosymbolic Learning: Characterization and Mitigating StrategiesEfthymia Tsamoura, Kaifu Wang, Dan Roth
Foundations of Top-k Decoding for Language ModelsGeorgy Noarov, Soham Mallick, Tao Wang, Sunay Joshi, Yan Sun, Yangxinyu Xie, Mengxin Yu, Edgar Dobriban
Probabilistic Stability Guarantees for Feature AttributionsHelen Jin, Anton Xue, Weiqiu You, Surbhi Goel, Eric Wong
MotionBind: Multi-modal Human Motion Alignment for Retrieval, Recognition, and GenerationKaleab A. Kinfu, Rene Vidal
Convergence Rates for Gradient Descent on the Edge of Stability for Overparameterized Least SquaresLachlan Ewen Macdonald, Hancheng Min, Leandro Palma, Salma Tarmoun, Ziqing Xu, Rene Vidal
TADA: Improved Diffusion Sampling with Training-free Augmented DynamicsTianrong Chen, Huangjie Zheng, David Berthelot, Jiatao Gu, Joshua M. Susskind, Shuangfei Zhai
Efficient PAC Learning for Realizable-statistic Models via Convex SurrogatesShivani Agarwal
Gradient Alignment in Physics-informed Neural Networks: A Second-order Optimization PerspectiveSifan Wang, Ananyae Kumar Bhartari, Bowen Li, Paris Perdikaris
Disentangling Misreporting from Genuine Adaptation in Strategic Settings: A Causal ApproachDylan Zapzalka, Trenton Chang, Lindsay Warrenburg, Sae-Hwan Park, Daniel K. Shenfeld, Ravi B. Parikh, Jenna Wiens, Maggie Makar
Once Upon an Input: Reasoning via Per-instance Program SynthesisAdam Stein, Neelay Velingker, Mayur Naik, Eric Wong
Neural Collapse Under Gradient Flow on Shallow ReLU Networks for Orthogonally Separable DataHancheng Min, Zhihui Zhu, Rene Vidal
BetaConform: Efficient MAP Estimation of LLM Ensemble Judgment Performance with Prior TransferHuaizhi Qu, Inyoung Choi, Zhen Tan, Song Wang, Sukwon Yun, Qi Long, Faizan Siddiqui, Kwonjoon Lee, Tianlong Chen
Conformal Information Pursuit for Interactively Guiding Large Language ModelsKwan Ho Ryan Chan, Yuyan Ge, Edgar Dobriban, Hamed Hassani, Rene Vidal
Deployment-efficient Reward-free Exploration with Linear Function ApproximationZihan Zhang, Yuxin Chen, Jason D. Lee, Simon Shaolei Du, Lin Yang, Ruosong Wang
CTSketch: Compositional Tensor Sketching for Scalable Neurosymbolic LearningSeewon Choi, Alaia Solko-Breslin, Rajeev Alur, Eric Wong
Covering Multiple Objectives with a Small Set of Solutions Using Bayesian OptimizationNatalie Maus, Kyurae Kim, Yimeng Zeng, Haydn Thomas Jones, Fangping Wan, Marcelo Der Torossian Torres, Cesar De La Fuente-Nunez, Jacob R. Gardner
Conformal Inference Under High-dimensional Covariate Shifts via Likelihood-ratio RegularizationSunay Joshi, Shayan Kiyani, George J. Pappas, Edgar Dobriban, Hamed Hassani
Nearly Dimension-independent Convergence of Mean-field Black-box Variational InferenceKyurae Kim, Yian Ma, Trevor Campbell, Jacob R. Gardner
PhysCtrl: Generative Physics for Controllable and Physics-grounded Video GenerationChen Wang, Chuhao Chen, Yiming Huang, Zhiyang Dou, Yuan Liu, Jiatao Gu, Lingjie Liu
Non-line-of-sight 3D Reconstruction with RadarHaowen Lai, Zitong Lan, Mingmin Zhao
Resounding Acoustic Fields with ReciprocityZitong Lan, Yiduo Hao, Mingmin Zhao
SECA: Semantically Equivalent and Coherent Attacks for Eliciting LLM HallucinationsBuyun Liang, Liangzu Peng, Jinqi Luo, Darshan Thaker, Kwan Ho Ryan Chan, Rene Vidal
On the Mechanisms of Weak-to-strong Generalization: A Theoretical PerspectiveBehrad Moniri, Hamed Hassani
A Dataset for Distilling Knowledge Priors from Literature for Therapeutic DesignHaydn Thomas Jones, Natalie Maus, Josh Magnus Ludan, Maggie Ziyu Huan, Jiaming Liang, Marcelo Der Torossian Torres, Jiatao Liang, Zachary Ives, Yoseph Barash, Cesar De La Fuente-Nunez, Jacob R. Gardner, Mark Yatskar
moPPIt-v3: Motif-specific Peptides Generated via Multi-objective-guided Discrete Flow MatchingTong Chen, Zachary Quinn, Yinuo Zhang, Pranam Chatterjee
High-Throughput Protein Perturbation Screens with AI-Designed DegradersLin Zhao, Aastha Pal, Tong Chen, Pranam Chatterjee
PrimateFace: A Resource for Generalizable Cross-Species Facial AnalysisFelipe Parodi, Konrad Kording
Unified Pretraining on Mixed Optophysiology and Electrophysiology Data Across Brain RegionsIan Jarratt Knight, Vinam Arora, Mehdi Azabou, Eva L. Dyer
A Scalable Self-Supervised Method for Modeling Human Intracranial Recordings During Natural BehaviorShivashriganesh P. Mahato, Jingyun Xiao, Alexandre Andre, Geeling Chau, Wenrui Ma, Ian Jarratt Knight, Duy Nguyen, Lawrence Jianqiao Hu, Bingni W. Brunton, Michael S. Beauchamp, Bijan Pesaran, Sergey A. Shuvaev, Eva L. Dyer
Predicting and Generating Antibiotics Against Future Pathogens with ApexOracleTianang Leng, Fangping Wan, Marcelo Der Torossian Torres, Cesar De La Fuente-Nunez
Token-Level Guided Discrete Diffusion for Membrane Protein DesignShrey Goel, Peregrine Michael Schray, Yinuo Zhang, Sophia Vincoff, Huong T. Kratochvil, Pranam Chatterjee
Explaining Temporal Effects in Sepsis PredictionChaehyeon Kim, Eric Wong
Instruction Following by Boosting Attention of Large Language ModelsVitoria Guardieiro, Adam Stein, Avishree Khare, Eric Wong
Where’s the Bug? Attention Probing for Scalable Fault LocalizationAdam Stein, Arthur Wayne, Aaditya Naik, Mayur Naik, Eric Wong
SuperActivators: Only the Tail of the Distribution Contains Reliable Concept SignalsCassandra Goldberg, Chaehyeon Kim, Adam Stein, Eric Wong
Learning from Frustration: Torsor CNNs on GraphsShreya Arya, Robert Ghrist, Daiyuan Li
The Binding Problem in Vision Models: Geometric, Functional, and Behavioral ApproachesLianghuan Huang, Yihao Li, Yingshan Chang, Saeed Salehi, Konrad Kording
Exploiting All Laplacian Eigenvectors for Node Classification with Graph TransformersVinam Arora, Divyansha Lachi, Shivashriganesh P. Mahato, Mehdi Azabou, Zihao Chen, Eva L. Dyer
Transferability of Graph Transformers with Convolutional Positional EncodingsJavier Porras-Valenzuela, Zhiyang Wang, Xiaotao Shang, Alejandro Ribeiro
GNN-Parametrized Diffusion Policies for Wireless Resource AllocationYigit Berkay Uslu, Samar Hadou, Shirin Saeedi Bidokhti, Alejandro Ribeiro
RELATE: A Schema-Agnostic Cross-Attention Encoder for Multimodal Relational GraphsJoe Meyer, Divyansha Lachi, Mahmoud Mohammadi, Roshan Reddy Upendra, Eva L. Dyer, Minghua Li, Tom Palczewski
Explaining Temporal Effects in Sepsis Prediction Chaehyeon Kim, Eric Wong
Integrating Slow Neural Oscillations and Physiological Burden for Trait Anxiety PredictionJungyoun Janice Min, Jiong Chen, Jingxuan Bao, Shu Yang, Yize Zhao, Li Shen, Duy Duong-Tran
ScooBDoob: Schrödinger Bridge with Doob’s h-transform for Molecular DynamicsYinuo Zhang, Sophia Tang, Pranam Chatterjee
Entangled Schrödinger Bridge MatchingSophia Tang, Yinuo Zhang, Pranam Chatterjee
Multi-objective Nanobody Design via Masked Discrete Diffusion with Simplex RefinementRosie Zhang, Pranam Chatterjee
Aalto University, Allen Institute, Amirkabir University of Technology, Arizona State University, Beijing Normal University, Beijing University of Posts and Telecommunications, Bernstein Center for Computational Neuroscience, Carnegie Mellon University, Center for AI Safety, Chinese University of Hong Kong, City University of Hong Kong, Columbia University, Dalian University of Technology, Donghua University, Shanghai, Emory University, Federal Institute of Ceará, Fudan University, Georgia Institute of Technology, Harvard University, Hong Kong University of Science and Technology, Huazhong University of Science and Technology, Ludwig-Maximilians-Universität München (University of Munich), Macau University of Science and Technology, McGill University, Mila – Quebec AI Institute, Nanyang Technological University, New Jersey Institute of Technology, New York University, Ohio State University, Peking University, Physical Intelligence, Princeton University, Rensselaer Polytechnic Institute, Shanghai AI Laboratory, Shanghai Jiao Tong University, Shanghaitech University, Southeast University, Stanford University, Sun Yat-sen University, Technion – Israel Institute of Technology, Technische Universität Berlin, Tel Aviv University, Texas A&M University – College Station, Université de Liège, Université de Montréal, University of British Columbia, University of California, Berkeley, University of California, Los Angeles, University of California, San Diego, University of Central Florida, University of Chicago, University of Electronic Science and Technology of China, University of Hong Kong, University of Illinois Urbana–Champaign, University of Macau, University of Michigan – Ann Arbor, University of North Carolina at Chapel Hill, University of Ottawa, University of Science and Technology of China, University of Toronto, University of Virginia, University of Washington, University of Wisconsin–Madison, Westlake University, Western Galilee College, Wuhan University, Yale University, Zhejiang University
Air Liquide, Amazon, Apple, Cerebras Systems, Inc., Google, Hengxin Technology Ltd., Honda, Huawei Technologies Ltd., Meta, Microsoft, Samsung, Stealth Mode Startup, Xiaomi Corporation
