12/23/2025 Megan Altmyer
Over 40 University of Illinois papers were selected for NeurIPS 2025. A preview poster event, Micro NeurIPS, was held at ECEB to provide the opportunity for faculty and students to share their work in advance of the international conference.
Written by Megan Altmyer
The Department of Electrical and Computer Engineering celebrated a significant milestone in artificial intelligence research during the 2025 NeurIPS cycle, with Illinois scholars showcasing their work both on campus and on the global stage.
In November, ECEB hosted Micro NeurIPS at Illinois, a preview poster event highlighting the more than 40 University of Illinois research papers accepted for presentation at the Thirty-Ninth Conference on Neural Information Processing Systems (NeurIPS 2025). The on-campus event brought together students, faculty, and researchers from across disciplines to explore Illinois contributions to one of the world’s premier AI conferences.
Micro NeurIPS served as an opportunity for presenters to share their work, receive feedback, and refine their posters ahead of the international conference. The event also highlighted the depth and breadth of Illinois research in areas such as machine learning theory, large-scale systems, computer vision, natural language processing, reinforcement learning, and responsible AI.
I love the work. It’s a lot of time and effort that’s been put in. It’s great to get to speak with students. A lot of friends too. They come by, and it’s really cool for them to understand what we’re working on.
-Will Shen
NeurIPS 2025 took place in December in San Diego, California, bringing together thousands of researchers from academia and industry. Widely regarded as one of the most influential conferences in artificial intelligence and machine learning, NeurIPS features highly selective paper acceptance and serves as a key forum for setting research directions across the field.
University of Illinois had a strong presence at the conference, with researchers presenting posters, participating in workshops, and engaging with the broader AI community. The number of accepted papers reflects both the strength of Illinois research and the collaborative culture that supports interdisciplinary innovation across campus.
Together, Micro NeurIPS at Illinois and NeurIPS 2025 underscore the department’s leadership in advancing AI research and preparing students to contribute at the highest levels of the field.
Illinois Papers Accepted to NeurIPS 2025:
More than 40 University of Illinois papers were accepted for presentation at NeurIPS 2025, spanning theory, algorithms, systems, and applications of artificial intelligence.
| 1 | Improving Data Efficiency for LLM Reinforcement Fine-tuning Through Difficulty-targeted Online Data Selection and Rollout Replay Yifan Sun, Huan Zhang |
LLM |
|---|---|---|
| 2 | AlphaOne: Reasoning Models Thinking Slow and Fast at Test Time (*workshop paper) Junyu Zhang, Han Wang, Huan Zhang |
LLM |
| 3 | RAST: Reasoning Activation in LLMs via Small-model Transfer Siru Ouyang, Xinyu Zhu, Zilin Xiao, Minhao Jiang, Yu Meng, Jiawei Han |
LLM |
| 4 | When Reasoning Meets Its Laws (*workshop paper) Junyu Zhang, Yifan Sun, Huan Zhang |
LLM |
| 5 | FGBench: A Dataset and Benchmark for Molecular Property Reasoning at Functional Group-Level in Large Language Models Xuan Liu, Siru Ouyang, Xianrui Zhong, Jiawei Han, Huimin Zhao |
LLM |
| 6 | Do LLMs Really Forget? Evaluating Unlearning with Knowledge Correlation and Confidence Awareness Peizhi Niu, Olgica Milenkovic |
LLM |
| 7 | AgMMU A Comprehensive Agricultural Multimodal Understanding and Reasoning Benchmark Ziqi Pang, Yunze Man, Yuxiong Wang |
LLM |
| 8 | LLM Strategic Reasoning: Agentic Study through Behavioral Game Theory. Jingru Jia, Zehua Yuan, Junhao Pan, Paul McNamara, Deming Chen |
LLM |
| 9 | MIRAGE: A Benchmark for Multimodal Information-Seeking and Reasoning in Agricultural Expert-Guided Conversations Vardhan Dongre, Chi Gui, Shubham Garg, Hooshang Nayyeri, Gokhan Tur, Dilek Hakkani-Tür, Vikram S. Adve |
LLM |
| 10 | Cost-Efficient LLM Training with Lifetime-Aware Tensor Offloading via GPUDirect Storage Ziqi Yuan, Haoyang Zhang, Yirui Eric Zhou, Jian Huang |
LLM, systems |
| 11 | Router-R1: Teaching LLMs Multi-Round Routing and Aggregation via Reinforcement Learning Haozhen Zhang, Tao Feng, Jiaxuan You |
LLM |
| 12 | Targeted Redirecting of Agentic Preferences Jehyeok Yeon, Hangoo Kang, Gagandeep Singh |
LLM |
| 13 | The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning Shivam Agarwal, Zimin Zhang, Lifan Yuan, Jiawei Han, Hao Peng |
LLM |
| 14 | DynamicRAG: Leveraging Outputs of Large Language Model as Feedback for Dynamic Reranking in Retrieval-Augmented Generation Jiashuo Sun, Xianrui Zhong, Sizhe Zhou, Jiawei Han |
LLM |
| 15 | DINGO: Constrained Inference for Diffusion LLMs Tarun Suresh, Debangshu Banerjee, Shubham Ugare, Sasa Misailovic, Gagandeep Singh |
LLM |
| 16 | Abstract Rendering: Computing All that is Seen in Gaussian Splat Scenes Chenxi Ji, Yangge Li, Xiangru Zhong, Huan Zhang, Sayan Mitra |
Vision |
| 17 | Can NeRFs “See” without Cameras? Chaitanya Amballa, Sattwik Basu, Yu-lin Wei, Romit Roy Choudhury |
Vision |
| 18 | Visual Sync: Multi‑Camera Synchronization via Cross‑View Object Motion Shaowei Liu, David Yao, Saurabh Gupta, Shenlong Wang |
Vision |
| 19 | CAR: Condition-Aware Reparameterization Aligns Source and Target for Better Flow Matching Chen Chen, Pengsheng Guo, Liangchen Song, Jiasen Lu, Rui Qian, Tsu-Jui Fu, Xinze Wang, Wei Liu, Yinfei Yang, Alex Schwing |
Vision |
| 20 | NoPo-Avatar: Generalizable and Animatable Avatars from Sparse Inputs without Human Poses Jing Wen, Alex Schwing, Shenlong Wang |
Vision |
| 21 | Fire360: A Benchmark for Robust Perception and Episodic Memory in Degraded 360° Firefighting Video Aditi Tiwari, Farzaneh Masoud, Dac Trong Nguyen, Jill Kraft, Heng Ji, Klara Nahrstedt |
Vision |
| 22 | REN: Fast and Efficient Region Encodings from Patch-Based Image Encoders Savya Khosla, Sethuraman T V, Barnett Lee, Alex Schwing, Derek Hoiem |
Vision |
| 23 | On Inductive Biases That Enable Generalization in Diffusion Transformers Jia An, De Wang, Pengsheng Guo, Jiebo Luo, Alex Schwing |
Vision |
| 24 | HoloScene: Simulation-Ready Interactive 3D Worlds from a Single Video Hongchi Xia, Chih-Hao Lin, Hao-Yu Hsu, Quentin Leboutet, Katelyn Gao, Michael Paulitsch, Benjamin Ummenhofer, Shenlong Wang |
Vision |
| 25 | MR. Video: MapReduce as an Effective Principle for Long Video Understanding Ziqi Pang, Yuxiong Wang |
Vision |
| 26 | RGB-Only Supervised Camera Parameter Optimization in Dynamic Scenes Fang Li, Hao Zhang, Narendra Ahuja |
Vision |
| 27 | Virtual Fitting Room: Generating Arbitrarily Long Videos of Virtual Try-On from a Single Image Junkun Chen, Yuxiong Wang |
Vision |
| 28 | One Token per Highly Selective Frame: Towards Extreme Compression for Long Video Understanding Zheyu Zhang, Ziqi Pang, Yuxiong Wang |
Vision |
| 29 | DiffEye: Diffusion-Based Continuous Eye-Tracking Data Generation Conditioned on Natural Images Ozgur Kara, Harris Nisar, James M. Rehg |
Vision, Generative |
| 30 | DMol: A Schedule-Driven Diffusion Model for Highly Efficient and Versatile Molecule Generation Peizhi Niu, Shane Wang, Olgica Milenkovic |
Generative |
| 31 | FalconWing: An Ultra-Light Fixed-Wing Platform for Indoor Aerial Applications Yan Miao, Will Shen, Hang Cui, Sayan Mitra (workshop paper) |
Vision |
| 32 | Model Context Protocol for Vision Agents: Schema, Memory, and World Model Implications Aditi Tiwari, Akshit Bhalla, Darshan Prasad |
Vision, AI Agents |
| 33 | GUI-Actor: Coordinate-Free Visual Grounding for GUI Agents Rui Yang, Huan Zhang |
AI Agents |
| 34 | Self-Guided Hierarchical Exploration for Generalist Foundation Model Web Agents Qianlan Yang, Yuxiong Wang |
AI Agents |
| 35 | Two‑Stage Learning of Stabilizing Neural Controllers via Zubov Sampling and Iterative Domain Expansion Xiangru Zhong, Haoyu Li, Bin Hu, Huan Zhang |
ML |
| 36 | Efficient Utility-Preserving Machine Unlearning Tianbai Yu |
ML |
| 37 | Hybrid Latent Reasoning via Reinforcement Learning Zhenrui Yue, Bowen Jin, Huimin Zeng, Honglei Zhuang, Zhen Qin, Jinsung Yoon, Lanyu Shang, Jiawei Han, Dong Wang |
RL |
| 38 | Sotopia-RL: Reward Design for Social Intelligence (workshop paper) Haofei Yu, Jiaxuan You |
RL |
| 39 | Scalable Policy-Based RL Algorithms for POMDPs Amey Anjarlekar, Rasoul Etesami, R. Srikant |
Theory, RL |
| 40 | Detection Is All You Need: A Feasible Optimal Prior-Free Black-Box Approach For Piecewise Stationary Bandits Argyrios Gerogiannis, Yu-Han Huang, Subhonmesh Bose, Venugopal V Veeravalli |
Theory |
| 41 | Riemannian Consistency Model Chaoran Cheng, Yusong Wang, Yuxin Chen, Xiangxin Zhou, Nanning Zheng, Ge Liu |
Theory |
| 42 | Sketched Adaptive Distributed Deep Learning: A Sharp Convergence Analysis Zhijie Chen, Qiaobo Li, Arindam Banerjee |
Theory |
| 43 | Distributionally Robust Performative Optimization Zhuangzhuang Jia, Roy Dong, Grani Hanasusanto |
Theory |
| 44 | Sketched Gaussian Mechanism for Private Federated Learning Qiaobo Li, Zhijie Chen, Arindam Banerjee |
Theory |
| 45 | Generative Caching for Structurally Similar Pts and Responses Sarthak Chakraborty, Suman Nath, Xuchao Zhang, Chetan Bansal, Indranil Gupta |
Infrastructure |
| 46 | Learning Reconfigurable Representations for Multimodal Federated Learning with Missing Data Duong Nguyen |
Federated learning |
| 47 | Clip-and-Verify: Linear Constraint-Driven Domain Clipping for Accelerating Neural Network Verification Duo Zhou, Grani Hanasusanto, Huan Zhang |
NN Verification |
| 48. | STRATUS: A Multi-agent System for Autonomous Reliability Engineering of Modern Clouds Yinfang Chen, Jackson Clark, Yiming Su, Tianyin Xu |
Reliability |