ICLR 2026 Oral Papers

ICLROral2026

All 224 papers accepted as Oral at ICLR 2026 — the top slice of the accepted programme. Every title links to its own page with authors, affiliation and sources.

All 224 papers

1. Common Corpus: The Largest Collection of Ethical Data for LLM Pre-TrainingPierre-Carl Langlais, Pavel Chizhov, Catherine Arnett et al. · Pleias
2. Q-RAG: Long Context Multi‑Step Retrieval via Value‑Based Embedder TrainingArtyom Sorokin, Nazar Buzun, Aleksandr Anokhin et al. · Applied AI Institute
3. From movement to cognitive maps: recurrent neural networks reveal how locomotor development shapes hippocampal spatial codingMarco P Abrate, Laurenz Muessig, Joshua P Bassett et al. · University College London, University of London
4. FIRE: Frobenius-Isometry Reinitialization for Balancing the Stability–Plasticity TradeoffIsaac Han, Sangyeon Park, Seungwon Oh et al. · Gwangju Institute of Science and Technology
5. Exchangeability of GNN Representations with Applications to Graph RetrievalKartik Nair, Indradyumna Roy, Soumen Chakrabarti et al. · Carnegie Mellon University
6. Enhancing Generative Auto-bidding with Offline Reward Evaluation and Policy SearchZhiyu Mou, Yiqin Lv, Miao Xu et al. · Alibaba Group
7. Why DPO is a Misspecified Estimator and How to Fix ItAditya Gopalan, Sayak Ray Chowdhury, Debangshu Banerjee · Indian Institute of Science
8. WebDevJudge: Evaluating (M)LLMs as Critiques for Web Development QualityChunyang Li, Yilun Zheng, Xinting Huang et al. · Department of Computer Science and Engineering, Hong Kong University of Science and Technology
9. SafeDPO: A Simple Approach to Direct Preference Optimization with Enhanced SafetyGeon-Hyeong Kim, Yu Jin Kim, Byoungjip Kim et al. · LG AI Research
10. MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data ScienceRan Xu, Yuchen Zhuang, Yishan Zhong et al. · Google DeepMind
11. Optimistic Task Inference for Behavior Foundation ModelsThomas Rupf, Marco Bagatella, Marin Vlastelica et al. · ETHZ - ETH Zurich
12. CounselBench: A Large-Scale Expert Evaluation and Adversarial Benchmarking of Large Language Models in Mental Health Question AnsweringYahan Li, Jifan Yao, John Bosco S. Bunyi et al.
13. AstaBench: Rigorous Benchmarking of AI Agents with a Scientific Research SuiteJonathan Bragg, Mike D'Arcy, Nishant Balepur et al. · Allen Institute for Artificial Intelligence
14. Neon: Negative Extrapolation From Self-Training Improves Image GenerationSina Alemohammad, Zhangyang Wang, Richard Baraniuk · University of Texas at Austin
15. Compositional Diffusion with Guided search for Long-Horizon PlanningUtkarsh Aashu Mishra, David He, Yongxin Chen et al. · Georgia Institute of Technology
16. Visual symbolic mechanisms: Emergent symbol processing in Vision Language ModelsRim Assouel, Declan Iain Campbell, Yoshua Bengio et al.
17. Addressing divergent representations from causal interventions on neural networksSatchel Grant, Simon Jerome Han, Alexa R. Tartaglini et al. · MATS
18. Cross-Domain Lossy Compression via Rate- and Classification-Constrained Optimal TransportNam Nguyen, Thinh Nguyen, Bella Bose · Oregon State University
19. Latent Fourier TransformMason Long Wang, Cheng-Zhi Anna Huang · Massachusetts Institute of Technology
20. GLASS Flows: Efficient Inference for Reward Alignment of Flow and Diffusion ModelsPeter Holderrieth, Uriel Singer, Tommi Jaakkola et al. · MIT
21. Latent Speech-Text TransformerYen-Ju Lu, Yashesh Gaur, Wei Zhou et al. · Johns Hopkins University
22. LoongRL: Reinforcement Learning for Advanced Reasoning over Long ContextsSiyuan Wang, Gaokai Zhang, Li Lyna Zhang et al. · Shanghai Jiaotong University
23. Speculative Actions: A Lossless Framework for Faster AI AgentsNaimeng Ye, Arnav Ahuja, Georgios Liargkovas et al. · Columbia University
24. Gaussian certified unlearning in high dimensions: A hypothesis testing approachAaradhya Pandey, Arnab Auddy, Haolin Zou et al. · Princeton University
25. RedTeamCUA: Realistic Adversarial Testing of Computer-Use Agents in Hybrid Web-OS EnvironmentsZeyi Liao, Jaylen Jones, Linxi Jiang et al. · Ohio State University, Columbus
26. Exploratory Causal Inference in SAEnceTommaso Mencattini, Riccardo Cadei, Francesco Locatello · EPFL - EPF Lausanne
27. Every Language Model Has a Forgery-Resistant SignatureMatthew Finlayson, Xiang Ren, Swabha Swayamdipta · University of Southern California
28. Pre-training under infinite computeKonwoo Kim, Suhas Kotha, Percy Liang et al.
29. GEPA: Reflective Prompt Evolution Can Outperform Reinforcement LearningLakshya A Agrawal, Shangyin Tan, Dilara Soylu et al. · University of California, Berkeley
30. p-less Sampling: A Robust Hyperparameter-Free Approach for LLM DecodingRunyan Tan, Shuang Wu, Phillip Howard · AI Labs, Thoughtworks
31. EigenBench: A Comparative Behavioral Measure of Value AlignmentJonathn Chang, Leonhard Piff, Suvadip Sana et al. · Cornell University
32. A Representer Theorem for Hawkes Processes via Penalized Least Squares MinimizationHideaki Kim, Tomoharu Iwata · NTT
33. UALM: Unified Audio Language Model for Understanding, Generation and ReasoningJinchuan Tian, Sang-gil Lee, Zhifeng Kong et al. · CMU, Carnegie Mellon University
34. Agent Data Protocol: Unifying Datasets for Diverse, Effective Fine-tuning of LLM AgentsYueqi Song, Ketan Ramaneti, Zaid Sheikh et al. · Carnegie Mellon University
35. OpenApps: Simulating Environment Variations to Measure UI Agent ReliabilityKaren Ullrich, Jingtong Su, Claudia Shi et al. · Meta AI
36. Overparametrization bends the landscape: BBP transitions at initialization in simple Neural NetworksBrandon Livio Annesi, Dario Bocchi, Chiara Cammarota · University of Roma "La Sapienza"
37. Softmax Transformers are Turing-CompleteHongjian Jiang, Michael Hahn, Georg Zetzsche et al. · Rheinland-Pfälzische Technische Universität
38. Steering the Herd: A Framework for LLM-based Control of Social LearningRaghu Arghal, Kevin He, Shirin Saeedi Bidokhti et al. · University of Pennsylvania, University of Pennsylvania
39. Quantitative Bounds for Length Generalization in TransformersZachary Izzo, Eshaan Nichani, Jason D. Lee · NEC Labs America
40. AgentGym-RL: An Open-Source Framework to Train LLM Agents for Long-Horizon Decision Making via Multi-Turn RLZhiheng Xi, Jixuan Huang, Chenyang Liao et al. · Fudan University
41. LLM Fingerprinting via Semantically Conditioned WatermarksThibaud Gloaguen, Robin Staab, Nikola Jovanović et al. · ETH Zurich
42. Watch your steps: Dormant Adversarial Behaviors that Activate upon LLM FinetuningThibaud Gloaguen, Mark Vero, Robin Staab et al. · ETH Zurich
43. Universal Inverse Distillation for Matching Models with Real-Data Supervision (No GANs)Nikita Maksimovich Kornilov, David Li, Tikhon Mavrin et al. · Applied AI Institute
44. Mixture-of-Experts Can Surpass Dense LLMs Under Strictly Equal ResourceHouyi Li, Ka Man Lo, Shijie Xuyang et al. · Fudan University
45. Pareto-Conditioned Diffusion Models for Offline Multi-Objective OptimizationJatan Shrestha, Santeri Heiskanen, Kari Hepola et al. · Aalto University
46. The Coverage Principle: How Pre-Training Enables Post-TrainingFan Chen, Audrey Huang, Noah Golowich et al. · Massachusetts Institute of Technology
47. Transformers are Inherently SuccinctPascal Bergsträßer, Ryan Cotterell, Anthony Widjaja Lin · Universität Kaiserslautern
48. Conformal Robustness Control: A New Strategy for Robust DecisionYang Hu, Jieren Tan, Changliang Zou et al. · Shanghai Jiaotong University
49. HATSolver: Learning Gröbner Bases with Hierarchical Attention TransformersMohamed Malhou, Ludovic Perret, Kristin E. Lauter
50. It's All Just Vectorization: einx, a Universal Notation for Tensor OperationsFlorian Fervers, Sebastian Bullinger, Christoph Bodensteiner et al. · Fraunhofer IOSB
51. Fast training of accurate physics-informed neural networks without gradient descentChinmay Datar, Taniya Kapoor, Abhishek Chandra et al. · Technische Universität München
52. In-Place Test-Time TrainingGuhao Feng, Shengjie Luo, Kai Hua et al. · Peking University
53. RAIN-Merging: A Gradient-Free Method to Enhance Instruction Following in Large Reasoning Models with Preserved Thinking FormatZhehao Huang, Yuhang Liu, Baijiong Lin et al. · Shanghai Jiaotong University
54. To Infinity and Beyond: Tool-Use Unlocks Length Generalization in State Space ModelsEran Malach, Omid Saremi, Sinead Williamson et al. · Apple
55. TileLang: Bridge Programmability and Performance in Modern Neural KernelsLei Wang, Yu Cheng, Yining Shi et al. · Peking University
56. Navigating the Latent Space Dynamics of Neural ModelsMarco Fumero, Luca Moschella, Emanuele Rodolà et al.
57. Non-Asymptotic Analysis of (Sticky) Track-and-StopRiccardo Poiani, Martino Bernasconi, Andrea Celli · Bocconi University
58. The Shape of Adversarial Influence: Characterizing LLM Latent Spaces with Persistent HomologyAideen Fay, Inés García-Redondo, Qiquan Wang et al. · Imperial College London
59. Hyperparameter Trajectory Inference with Conditional Lagrangian Optimal TransportHarry Amad, Mihaela van der Schaar · University of Cambridge
60. Vid-LLM: A Compact Video-based 3D Multimodal LLM with Reconstruction–Reasoning SynergyHaijier Chen, Bo Xu, Shoujian zhang et al. · Wuhan University
61. From Markov to Laplace: How Mamba In-Context Learns Markov ChainsMarco Bondaschi, Nived Rajaraman, Xiuying Wei et al. · EPFL - EPF Lausanne
62. Huxley-G\"odel Machine: Human-Level Coding Agent Development by an Approximation of the Optimal Self-Improving MachineWenyi Wang, Piotr Piękos, Li Nanbo et al. · King Abdullah University of Science and Technology
63. Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous EnvironmentsRomain Froger, Pierre Andrews, Matteo Bettini et al. · Facebook
64. Fast Escape, Slow Convergence: Learning Dynamics of Phase Retrieval under Power-Law DataGuillaume Braun, Bruno Loureiro, Minh Ha Quang et al. · RIKEN
65. AutoEP: LLMs-Driven Automation of Hyperparameter Evolution for Metaheuristic AlgorithmsZhenxing Xu, Yizhe Zhang, Weidong Bao et al. · National University of Defense Technology
66. On the Wasserstein Geodesic Principal Component Analysis of probability measuresNina Vesseron, Elsa Cazelles, Alice Le Brigant et al. · Ecole Nationale de la Statistique et de l'Administration Economique
67. Discount Model Search for Quality Diversity Optimization in High-Dimensional Measure SpacesBryon Tjanaka, Henry Chen, Matthew Christopher Fontaine et al. · University of Southern California
68. Decentralized Attention Fails Centralized Signals: Rethinking Transformers for Medical Time SeriesGuoqi Yu, Juncheng Wang, Chen Yang et al.
69. Why Low-Precision Transformer Training Fails: An Analysis on Flash AttentionHaiquan Qiu, Quanming Yao · Tsinghua University, Tsinghua University
70. Overcoming Joint Intractability with Lossless Hierarchical Speculative DecodingYuxuan Zhou, Fei Huang, Heng Li et al. · Baidu
71. In-the-Flow Agentic System Optimization for Effective Planning and Tool UseZhuofeng Li, Haoxiang Zhang, Seungju Han et al. · Texas A&M University - College Station
72. What's In My Human Feedback? Learning Interpretable Descriptions of Preference DataRajiv Movva, Smitha Milli, Sewon Min et al. · University of California, Berkeley
73. SimuHome: A Temporal- and Environment-Aware Benchmark for Smart Home LLM AgentsGyuhyeon Seo, Jungwoo Yang, Junseong Pyo et al. · Seoul National University
74. Reasoning with Sampling: Your Base Model is Smarter Than You ThinkAayush Karan, Yilun Du · Harvard University
75. Learning to Segment for Vehicle Routing ProblemsWenbin Ouyang, Sirui Li, Yining Ma et al. · Massachusetts Institute of Technology
76. WAVE: Learning Unified & Versatile Audio-Visual Embeddings with Multimodal LLMChangli Tang, Qinfan Xiao, Ke Mei et al. · Tsinghua University, Tsinghua University
77. mCLM: A Modular Chemical Language Model that Generates Functional and Makeable MoleculesCarl Edwards, Chi Han, Gawon Lee et al. · Genentech
78. Seeing Through the Brain: New Insights from Decoding Visual Stimuli with fMRIZheng Huang, Enpei Zhang, Weikang Qiu et al.
79. Temporal Sparse Autoencoders: Leveraging the Sequential Nature of Language for InterpretabilityUsha Bhalla, Alex Oesterling, Claudio Mayrink Verdun et al.
80. VibeVoice: Expressive Podcast Generation with Next-Token DiffusionZhiliang Peng, Jianwei Yu, Wenhui Wang et al. · Microsoft
81. AdAEM: An Adaptively and Automated Extensible Measurement of LLMs' Value DifferenceJing Yao, Shitong Duan, Xiaoyuan Yi et al. · Microsoft
82. Omni-Reward: Towards Generalist Omni-Modal Reward Modeling with Free-Form PreferencesZhuoran Jin, Hongbang Yuan, Kejian Zhu et al. · Institute of Automation, Chinese Academy of Sciences
83. Spherical Watermark: Encryption-Free, Lossless Watermarking for Diffusion ModelsXiaoxiao Hu, Jiaqi Jin, Sheng Li et al. · Fudan University
84. Rodrigues Network for Learning Robot ActionsJialiang Zhang, Haoran Geng, Yang You et al. · Peking University
85. CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at ScaleZhun Wang, Tianneng Shi, Jingxuan He et al.
86. Energy-Based Transformers are Scalable Learners and ThinkersAlexi Gladstone, Ganesh Nanduru, Md Mofijul Islam et al. · Flapping Airplanes
87. Global Resolution: Optimal Multi-Draft Speculative Sampling via Convex OptimizationRahul Krishna Thomas, Arka Pal · Columbia University
88. FALCON: Few-step Accurate Likelihoods for Continuous FlowsDanyal Rehman, Tara Akhound-Sadegh, Artem Gazizov et al. · Mila - Quebec Artificial Intelligence Institute
89. How Reliable is Language Model Micro-Benchmarking?Gregory Yauney, Shahzaib Saqib Warraich, Swabha Swayamdipta · Amazon
90. EDIT-Bench: Evaluating LLM Abilities to Perform Real-World Instructed Code EditsWayne Chi, Valerie Chen, Ryan Shar et al. · Carnegie Mellon University
91. Online Learning and Equilibrium Computation with Ranking FeedbackMingyang Liu, Yongshan Chen, Zhiyuan Fan et al. · Massachusetts Institute of Technology
92. MC-Search: Evaluating and Enhancing Multimodal Agentic Search with Structured Long Reasoning ChainsXuying Ning, Dongqi Fu, Tianxin Wei et al. · University of Illinois at Urbana-Champaign
93. True Self-Supervised Novel View Synthesis is TransferableThomas Mitchel, Hyunwoo Ryu, Vincent Sitzmann
94. Revela: Dense Retriever Learning via Language ModelingFengyu Cai, Tong Chen, Xinran Zhao et al. · Technische Universität Darmstadt
95. LLMs Get Lost In Multi-Turn ConversationPhilippe Laban, Hiroaki Hayashi, Yingbo Zhou et al. · Microsoft
96. TTSDS2: Resources and Benchmark for Evaluating Human-Quality Text to Speech SystemsChristoph Minixhofer, Ondrej Klejch, Peter Bell · University of Edinburgh, University of Edinburgh
97. Mamba-3: Improved Sequence Modeling using State Space PrinciplesAakash Lahoti, Kevin Li, Berlin Chen et al. · CMU, Carnegie Mellon University
98. Is it Thinking or Cheating? Detecting Implicit Reward Hacking by Measuring Reasoning EffortXinpeng Wang, Nitish Joshi, Barbara Plank et al. · Ludwig-Maximilians-Universität München
99. Causal Structure Learning in Hawkes Processes with Complex Latent Confounder NetworksSongyao Jin, Biwei Huang · University of California, San Diego
100. Mean Flow Policy with Instantaneous Velocity Constraint for One-step Action GenerationGuojian Zhan, Letian Tao, Pengcheng Wang et al. · Tsinghua University
101. Radiometrically Consistent Gaussian Surfels for Inverse RenderingKyu Beom Han, Jaeyoon Kim, Woo Jae Kim et al. · Korea Advanced Institute of Science & Technology
102. Hubble: a Model Suite to Advance the Study of LLM MemorizationJohnny Wei, Ameya Godbole, Mohammad Aflah Khan et al. · University of Southern California
103. Differentially Private Domain DiscoveryVinod Raman, Travis Dick, Matthew Joseph · Google DeepMind
104. CauKer: Classification Time Series Foundation Models Can Be Pretrained on Synthetic DataShifeng Xie, Vasilii Feofanov, Jianfeng Zhang et al. · Télécom Paris
105. Non-Convex Federated Optimization under Cost-Aware Client SelectionXiaowen Jiang, Anton Rodomanov, Sebastian U Stich · CISPA Helmholtz Center for Information Security
106. Verifying Chain-of-Thought Reasoning via Its Computational GraphZheng Zhao, Yeskendir Koishekenov, Xianjun Yang et al. · University of Edinburgh, University of Edinburgh
107. How Learning Rate Decay Wastes Your Best Data in Curriculum-Based LLM PretrainingKairong Luo, Zhenbo Sun, Haodong Wen et al. · Tsinghua University
108. Probabilistic Kernel Function for Fast Angle TestingKejing Lu, Chuan Xiao, Yoshiharu Ishikawa · Yamanashi University
109. RefineStat: Efficient Exploration for Probabilistic Program SynthesisMadhav Kanda, Shubham Ugare, Sasa Misailovic · University of Illinois at Urbana-Champaign
110. Scaling Atomistic Protein Binder Design with Generative Pretraining and Test-Time ComputeKieran Didi, Zuobai Zhang, Guoqing Zhou et al. · NVIDIA
111. Structured Flow Autoencoders: Learning Structured Probabilistic Representations with Flow MatchingYidan Xu, Yixin Wang, XuanLong Nguyen
112. Plug-and-Play Compositionality for Boosting Continual Learning with Foundation ModelsWeiduo Liao, Fei Han, Hisao Ishibuchi et al. · City University of Hong Kong
113. Generating metamers of human scene understandingRitik Raina, Abe Leite, Alexandros Graikos et al. · State University of New York at Stony Brook
114. Efficient Resource-Constrained Training of Transformers via Subspace OptimizationLe-Trung Nguyen, Enzo Tartaglione, Van-Tam Nguyen · Télécom Paris
115. Learning with Dual-level Noisy Correspondence for Multi-modal Entity AlignmentHaobin Li, Yijie Lin, Peng Hu et al. · School of Software Engineering, Sichuan University, Sichuan University
116. Pinet: Optimizing hard-constrained neural networks with orthogonal projection layersPanagiotis D. Grontas, Antonio Terpin, Efe C. Balta et al. · International Business Machines
117. ParaRNN: Unlocking Parallel Training of Nonlinear RNNs for Large Language ModelsFederico Danieli, Pau Rodriguez, Miguel Sarabia et al. · Apple
118. Exploring Synthesizable Chemical Space with Iterative Pathway RefinementsSeul Lee, Karsten Kreis, Srimukh Prasad Veccham et al. · Korea Advanced Institute of Science and Technology
119. On The Surprising Effectiveness of a Single Global Merging in Decentralized LearningTongtian Zhu, Tianyu Zhang, Mingze Wang et al. · Zhejiang University
120. Visual Planning: Let's Think Only with ImagesYi Xu, Chengzu Li, Han Zhou et al. · Mistral AI
121. Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning TasksTaishi Nakamura, Satoki Ishikawa, Masaki Kawamura et al. · Institute of Science Tokyo
122. Exploratory Diffusion Model for Unsupervised Reinforcement LearningChengyang Ying, Huayu Chen, Xinning Zhou et al. · Tsinghua University, Tsinghua University
123. DCFold: Efficient Protein Structure Generation with Single Forward PassZhe Zhang, Yuanning Feng, Yuxuan Song et al. · Tsinghua University
124. Triple-BERT: Do We Really Need MARL for Order Dispatch on Ride-Sharing Platforms?Zijian Zhao, Sen Li · Huawei Technologies Ltd.
125. InfoNCE Induces Gaussian DistributionRoy Betser, Eyal Gofer, Meir Yossef Levi et al. · Technion - Israel Institute of Technology, Technion
126. Coupling Experts and Routers in Mixture-of-Experts via an Auxiliary LossAng Lv, Jin Ma, Yiyuan Ma et al. · Renmin University of China
127. P-GenRM: Personalized Generative Reward Model with Test-time User-based ScalingPinyi Zhang, Ting-En Lin, Yuchuan Wu et al. · East China Normal University
128. Modality-free Graph In-context AlignmentWei Zhuo, Siqiang Luo · Nanyang Technological University
129. TRACE: Your Diffusion Model is Secretly an Instance Edge DetectorSanghyun Jo, Ziseok Lee, Wooyeol Lee et al. · OGQ
130. Scaling Laws and Spectra of Shallow Neural Networks in the Feature Learning RegimeLeonardo Defilippis, Yizhou Xu, Julius Girardin et al. · Ecole Normale Supérieure, Ecole Normale Supérieure de Paris
131. MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory AgentHongli Yu, Tinghong Chen, Jiangtao Feng et al. · Tsinghua University
132. TD-JEPA: Latent-predictive Representations for Zero-Shot Reinforcement LearningMarco Bagatella, Matteo Pirotta, Ahmed Touati et al. · Max Planck Institute for Intelligent Systems, Max Planck Institute for Intelligent Systems
133. NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at ScaleChunrui Han, Guopeng Li, Jingwei Wu et al. · Stepfun
134. Differentiable Model Predictive Control on the GPUEmre Adabag, Marcus Greiff, John Subosits et al. · University of Michigan - Ann Arbor
135. Planner Aware Path Learning in Diffusion Language Models TrainingFred Zhangzhi Peng, Zachary Bezemek, Jarrid Rector-Brooks et al. · Duke University
136. PateGAIL++: Utility Optimized Private Trajectory Generation with Imitation LearningYingjie Ma, Bijal Bharadva, Xin Zhang et al.
137. MetaEmbed: Scaling Multimodal Retrieval at Test-Time with Flexible Late InteractionZilin Xiao, Qi Ma, Mengting Gu et al. · Rice University
138. Monocular Normal Estimation via Shading Sequence EstimationZongrui Li, Xinhua Ma, Minghui Hu et al. · Nanyang Technological University
139. Shoot First, Ask Questions Later? Building Rational Agents that Explore and Act Like PeopleGabriel Grand, Valerio Pepe, Joshua B. Tenenbaum et al. · Massachusetts Institute of Technology
140. Intrinsic Entropy of Context Length Scaling in LLMsJingzhe Shi, Qinwei Ma, Hongyi Liu et al. · Tsinghua University
141. Latent Particle World Models: Self-supervised Object-centric Stochastic Dynamics ModelingTal Daniel, Carl Qi, Dan Haramati et al. · Carnegie Mellon University
142. Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign PromptsZhaomin Wu, Mingzhe Du, See-Kiong Ng et al. · National University of Singapore
143. LLM DNA: Tracing Model Evolution via Functional RepresentationsZhaomin Wu, Haodong Zhao, Ziyang Wang et al. · National University of Singapore
144. Generative Universal Verifier as Multimodal Meta-ReasonerXinchen Zhang, Xiaoying Zhang, Youbin Wu et al. · ByteDance Seed
145. $PhyWorldBench$: A Comprehensive Evaluation of Physical Realism in Text-to-Video ModelsJing Gu, Xian Liu, Yu Zeng et al. · xAI
146. World-In-World: World Models in a Closed-Loop WorldJiahan Zhang, Muqing Jiang, Nanru Dai et al.
147. Multimodal Aligned Semantic Knowledge for Unpaired Image-text MatchingLaiguo Yin, Yixin Zhang, YUQING SUN et al. · Shandong University
148. DiffusionNFT: Online Diffusion Reinforcement with Forward ProcessKaiwen Zheng, Huayu Chen, Haotian Ye et al. · Tsinghua University
149. The Spacetime of Diffusion Models: An Information Geometry PerspectiveRafal Karczewski, Markus Heinonen, Alison Pouplin et al. · Aalto University
150. Reducing Belief Deviation in Reinforcement Learning for Active Reasoning of LLM AgentsDeyu Zou, Yongqiang Chen, Jianxiang Wang et al. · Department of Computer Science and Engineering, The Chinese University of Hong Kong
151. Information Shapes Koopman RepresentationXiaoyuan Cheng, Wenxuan Yuan, Yiming Yang et al. · University College London, University of London
152. Sequences of Logits Reveal the Low Rank Structure of Language ModelsNoah Golowich, Allen Liu, Abhishek Shetty · Microsoft
153. BIRD-INTERACT: Re-imagining Text-to-SQL Evaluation via Lens of Dynamic InteractionsNan Huo, Xiaohan Xu, Jinyang Li et al. · the University of Hong Kong, University of Hong Kong
154. EditVerse: Unifying Image and Video Editing and Generation with In-Context LearningXuan Ju, Tianyu Wang, Yuqian Zhou et al. · Chinese University of Hong Kong
155. The Art of Scaling Reinforcement Learning Compute for LLMsFnu Devvrit, Lovish Madaan, Rishabh Tiwari et al. · , University of Texas at Austin
156. Partition Generative Modeling: Masked Modeling Without MasksJustin Deschenaux, Lan Tran, Caglar Gulcehre · EPFL - EPF Lausanne
157. How Do Transformers Learn to Associate Tokens: Gradient Leading Terms Bring Mechanistic InterpretabilityShawn Im, Changdae Oh, Zhen Fang et al. · Department of Computer Science, University of Wisconsin - Madison
158. On the Reasoning Abilities of Masked Diffusion Language ModelsAnej Svete, Ashish Sabharwal · Department of Computer Science, ETHZ - ETH Zurich
159. FRABench and UFEval: Unified Fine-grained Evaluation with Task and Aspect GeneralizationShibo Hong, Jiahao Ying, Haiyuan Liang et al. · Fudan University
160. Quotient-Space Diffusion ModelsYixian Xu, Yusong Wang, Shengjie Luo et al. · Peking University
161. SWINGARENA: Adversarial Programming Arena for Long-context GitHub Issue SolvingWendong XU, Jing Xiong, Chenyang Zhao et al. · University of Hong Kong
162. AnyUp: Universal Feature UpsamplingThomas Wimmer, Prune Truong, Marie-Julie Rakotosaona et al. · Max-Planck Institute
163. High-dimensional Analysis of Synthetic Data SelectionParham Rezaei, Filip Kovačević, Francesco Locatello et al. · Institute of Science and Technology Austria
164. Task-free Adaptive Meta Black-box OptimizationChao Wang, Licheng Jiao, Lingling Li et al. · Xidian University
165. Token-Importance Guided Direct Preference OptimizationNing Yang, Hai Lin, Yibo Liu et al. · Institute of automation, Chinese academy of science, Chinese Academy of Sciences
166. DTO-KD: Dynamic Trade-off Optimization for Effective Knowledge DistillationZeeshan Hayder, Ali Cheraghian, Lars Petersson et al. · Google
167. Distributional Equivalence in Linear Non-Gaussian Latent-Variable Cyclic Causal Models: Characterization and LearningHaoyue Dai, Immanuel Albrecht, Peter Spirtes et al. · Carnegie Mellon University
168. Reliable Weak-to-Strong Monitoring of LLM AgentsNeil Kale, Chen Bo Calvin Zhang, Kevin Zhu et al. · School of Computer Science, Carnegie Mellon University
169. Multiplayer Nash Preference OptimizationFang Wu, Xu Huang, Weihao Xuan et al.
170. Multi-Domain Riemannian Graph Gluing for Building Graph Foundation ModelsLi Sun, Zhenhao Huang, Silei Chen et al. · Beijing University of Post and Telecommunications
171. TROLL: Trust Regions Improve Reinforcement Learning for Large Language ModelsPhilipp Becker, Niklas Freymuth, Serge Thilges et al. · Facebook
172. FlashVID: Efficient Video Large Language Models via Training-free Tree-based Spatiotemporal Token MergingZiyang Fan, Keyu Chen, Ruilong Xing et al. · Harbin Institute of Technology, Shenzhen
173. WSM: Decay-Free Learning Rate Schedule via Checkpoint Merging for LLM Pre-trainingChangxin Tian, jiapeng wang, Qian Zhao et al. · Ant Group
174. MrRoPE: Mixed-radix Rotary Position EmbeddingQingyuan Tian, Wenhong Zhu, Xiaoran Liu et al. · Shanghai Jiaotong University
175. OpenThoughts: Data Recipes for Reasoning ModelsEtash Kumar Guha, Ryan Marten, Sedrick Keh et al. · RIKEN
176. Text-to-3D by Stitching a Multi-view Reconstruction Network to a Video GeneratorHyojun Go, Dominik Narnhofer, Goutam Bhat et al. · ETHZ - ETH Zurich
177. A Scalable Distributed Framework for Multimodal GigaVoxel Image RegistrationRohit Jena, Vedant Zope, Pratik Chaudhari et al. · University of Pennsylvania
178. Invisible Safety Threat: Malicious Finetuning for LLM via SteganographyGuangnian Wan, Xinyin Ma, Gongfan Fang et al. · National University of Singapore
179. Temporal superposition and feature geometry of RNNs under memory demandsPratyaksh Sharma, Alexandra Maria Proca, Lucas Prieto et al. · G-Research
180. InfoTok: Adaptive Discrete Video Tokenizer via Information-Theoretic CompressionHaotian Ye, Qiyuan He, Jiaqi Han et al. · Stanford University
181. Actions Speak Louder than Prompts: A Large-Scale Study of LLMs for Graph InferenceBen Finkelshtein, Silviu Cucerzan, Sujay Kumar Jauhar et al. · University of Oxford
182. Through the Lens of Contrast: Self-Improving Visual Reasoning in VLMsZhiyu Pan, Yizheng Wu, Jiashen Hua et al. · Alibaba Group
183. SAFETY-GUIDED FLOW (SGF): A UNIFIED FRAMEWORK FOR NEGATIVE GUIDANCE IN SAFE GENERATIONMingyu Kim, Young-Heon Kim, Mijung Park · Kookmin University
184. Depth Anything 3: Recovering the Visual Space from Any ViewsHaotong Lin, Sili Chen, Jun Hao Liew et al. · Zhejiang University
185. Taming Momentum: Rethinking Optimizer States Through Low-Rank ApproximationZhengbo Wang, Jian Liang, Ran He et al. · University of Science and Technology of China
186. Premise Selection for a Lean HammerThomas Zhu, Joshua Clune, Jeremy Avigad et al. · ByteDance Inc.
187. DepthLM: Metric Depth from Vision Language ModelsZhipeng Cai, Ching-Feng Yeh, Hu Xu et al. · Meta
188. Hallucination Begins Where Saliency DropsXiaofeng Zhang, Yuanchao Zhu, Chaochen Gu et al. · Shanghai Jiao Tong University
189. Overthinking Reduction with Decoupled Rewards and Curriculum Data SchedulingShuyang Jiang, Yusheng Liao, Ya Zhang et al. · Fudan University
190. On the Generalization Capacities of MLLMs for Spatial IntelligenceGongjie Zhang, Wenhao Li, Quanhao Qian et al. · Alibaba Group
191. The Polar Express: Optimal Matrix Sign Methods and their Application to the Muon AlgorithmNoah Amsel, David Persson, Christopher Musco et al. · NYU, New York University
192. Characterizing the Discrete Geometry of ReLU NetworksBlake B. Gaines, Jinbo Bi · University of Connecticut
193. ThinKV: Thought-Adaptive KV Cache Compression for Efficient Reasoning ModelsAkshat Ramachandran, Marina Neseem, Charbel Sakr et al. · Georgia Institute of Technology
194. Difficult Examples Hurt Unsupervised Contrastive Learning: A Theoretical PerspectiveYi-Ge Zhang, Jingyi Cui, Qiran Li et al. · Peking University
195. Let Features Decide Their Own Solvers: Hybrid Feature Caching for Diffusion TransformersShikang Zheng, Guantao Chen, Qinming Zhou et al.
196. MotionStream: Real-Time Video Generation with Interactive Motion ControlsJoonghyuk Shin, Zhengqi Li, Richard Zhang et al. · Seoul National University
197. BioX-Bridge: Model Bridging for Unsupervised Cross-Modal Knowledge Transfer across BiosignalsChenqi Li, Yu Liu, Timothy Denison et al. · University of Oxford
198. Compactness and Consistency: A Conjoint Framework for Deep Graph ClusteringWei Ju, Siyu Yi, Kangjie Zheng et al.
199. WAFT: Warping-Alone Field Transforms for Optical FlowYihan Wang, Jia Deng · Princeton University
200. FlashWorld: High-quality 3D Scene Generation within SecondsXinyang Li, Tengfei Wang, Zixiao Gu et al. · Xiamen University
201. TabStruct: Measuring Structural Fidelity of Tabular DataXiangjian Jiang, Nikola Simidjievski, Mateja Jamnik · University of Cambridge
202. Uncover Underlying Correspondence for Robust Multi-view ClusteringHaochen Zhou, Guofeng Ding, Mouxing Yang et al. · Sichuan University
203. Instilling an Active Mind in Avatars via Cognitive SimulationJianwen Jiang, Weihong Zeng, Zerong Zheng et al.
204. Diffusion Language Model Knows the Answer Before It DecodesPengxiang Li, Yefan Zhou, Dilxat Muhtar et al. · Hong Kong Polytechnic University
205. Generative Human Geometry DistributionXiangjun Tang, Biao Zhang, Peter Wonka · King Abdullah University of Science and Technology
206. EmotionThinker: Prosody-Aware Reinforcement Learning for Explainable Speech Emotion ReasoningDingdong WANG, Shujie LIU, Tianhua Zhang et al. · Chinese University of Hong Kong, The Chinese University of Hong Kong
207. Mastering Sparse CUDA Generation through Pretrained Models and Deep Reinforcement LearningYaoyu Wang, Hankun Dai, Zhidong Yang et al. · University of Chinese Academy of Sciences
208. LongWriter-Zero: Mastering Ultra-Long Text Generation via Reinforcement LearningYuhao Wu, Yushi Bai, Zhiqiang Hu et al. · Singapore University of Technology and Design
209. RealPDEBench: A Benchmark for Complex Physical Systems with Real-World DataPeiyan Hu, Haodong Feng, Hongyuan Liu et al. · Chinese Academy of Sciences
210. Improving Diffusion Models for Class-imbalanced Training Data via Capacity ManipulationFeng Hong, Jiangchao Yao, Yifei Shen et al. · Shanghai Jiao Tong University
211. SANA-Video: Efficient Video Generation with Block Linear Diffusion TransformerJunsong Chen, Yuyang Zhao, Jincheng YU et al. · University of Hong Kong
212. Stable Video Infinity: Infinite-Length Video Generation with Error RecyclingWuyang Li, Wentao Pan, Po-Chien Luan et al. · EPFL - EPF Lausanne
213. Extending Sequence Length is Not All You Need: Effective Integration of Multimodal Signals for Gene Expression PredictionZhao Yang, Yi Duan, Jiwei Zhu et al. · Renmin University of China
214. MomaGraph: State-Aware Unified Scene Graphs with Vision-Language Models for Embodied Task PlanningYuanchen Ju, Yongyuan Liang, Yen-Jen Wang et al.
215. Learning to See Before Seeing: Demystifying LLM Visual Priors from Language Pre-trainingJunlin Han, Shengbang Tong, David Fan et al. · University of Oxford
216. Benchmarking Empirical Privacy Protection for Adaptations of Large Language ModelsBartłomiej Marek, Lorenzo Rossi, Vincent Hanke et al. · CISPA Helmholtz Center for Information Security
217. ScaleCUA: Scaling Open-Source Computer Use Agents with Cross-Platform DataZhaoyang Liu, JingJing Xie, Zichen Ding et al. · Hong Kong University of Science and Technology
218. Train-before-Test Harmonizes Language Model RankingsGuanhua Zhang, Ricardo Dominguez-Olmedo, Moritz Hardt · Max Planck Institute for Intelligent Systems, Max-Planck Institute
219. One for Two: A Unified Framework for Imbalanced Graph Classification via Dynamic Balanced PrototypeGuanjun Wang, Binwu Wang, Jiaming Ma et al. · University of Science and Technology of China
220. Veritas: Generalizable Deepfake Detection via Pattern-Aware ReasoningHao Tan, jun lan, Zichang Tan et al. · Institute of automation, Chinese Academy of Sciences
221. Semi-Supervised Preference Optimization with Limited FeedbackSeonggyun Lee, Sungjun Lim, Seojin Park et al. · Yonsei University
222. Locality-aware Parallel Decoding for Efficient Autoregressive Image GenerationZhuoyang Zhang, Luke J. Huang, Chengyue Wu et al. · Massachusetts Institute of Technology
223. Reasoning as Representation: Rethinking Visual Reinforcement Learning in Image Quality AssessmentShijie Zhao, Xuanyu Zhang, Weiqi Li et al. · ByteDance Inc.
224. Half-order Fine-Tuning for Diffusion Model: A Recursive Likelihood Ratio OptimizerTao Ren, Zishi Zhang, Jinyang Jiang et al. · Peking University

← Browse the whole archive