Infinities

Follow this topic Ask about this topic

Momentum

3 papers in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 28

All topics

One from Infinity: Actualizing Futures from Pretrained World Models into Robot Actions
Sep 29, 2026Bang Du, Yichen Xie, Shuqi Zhao +3Efficient World-Action Model Video World Models
University of California, Berkeley · Southern University of Science and Technology · Xi’an Jiaotong University
Quasi Linear Kernel Attention with Infinite Capacity
Sep 28, 2026Nicolaj Rux, Johannes Hertrich, Sebastian NeumayerKimi Delta Attention Kernel Method
Faculty of Mathematics, Chemnitz University of Technology, 09126 Chemnitz, Germany · Institute of Computer Science, University of Göttingen, 37073 Göttingen, Germany
Critical initialization destabilizes higher input derivatives in wide scalar-input networks
Sep 10, 2026Prashant Singh, Pranav SinghTwo-Layer Neural Networks Mean-Field Limit
Department of Mathematics, Indian Institute of Technology Ropar, Rupnagar, Punjab, 140001, India
Correlated initialization of deep residual networks
Sep 3, 2026Felix Benning, Ivan Nourdin, Giovanni PeccatiResidual Networks Infinities
A Quantum Roadmap for Softmax Attention: Exact Born-Rule Analogs for Softmax Attention on the Probability Simplex
Aug 11, 2026Eric A. F. Reinhardt, Adam J. HauserGumbel-Softmax Relaxation Qubit
Department of Physics and Astronomy University of Alabama, Tuscaloosa, AL 35487, USA
Voting Method Synthesis on an Infinite Domain: A Possibility Theorem for Positive Involvement
Aug 9, 2026Wesley H. HollidaySatisfiability Majority Voting
University of California, Berkeley
Correlation flow governs learning at criticality
Aug 8, 2026Andrea Combette, Nelly Pustelnik, Antoine VenailleInfinities Deep Learning
CNRS, ENS de Lyon, Laboratoire de Physique UMR5672, Lyon, France
Tired Actor: Fatigue-Informed Character Control
Aug 4, 2026Shengyuan Zhang, Xinpeng Liu, Muchun Niu +5Motion Imitation Infinities
Shanghai Jiao Tong University · Shanghai Innovation Institute · Harbin Institute of Technology
On the robustness of noisy solutions in non-convex neural networks
Jul 29, 2026Enrico M. Malatesta, Alessandra Passalacqua, Riccardo ZecchinaSlow Relaxation Modes Infinities
Department of Computing Sciences, Bocconi University, Milano, Italy · Bocconi Institute for Data Science and Analytics (BIDSA)
Effective Parameters, Real Behavior: Renormalization for Robotics -- From Infinite Electron Mass to Sim-to-Real Gap
Jul 27, 2026Youran Sun, Jiaxuan Guo, Xingyu Ren +2Sim-To-Real Gap Renormalization Group
University of Maryland, College Park · Stanford University · The Chinese University of Hong Kong
When Can Depth Replace Precision? A Resource Theory of Quantized Neural Computation
Jul 25, 2026Mojtaba SoltanalianFull-Precision Performance Circuit Depth
Department of Electrical and Computer Engineering University of Illinois Chicago 851 S. Morgan St., MC 154 Chicago, IL 60607, USA
Infinite Canons: Maximally Self-Similar Melodic Lines and Canons with Infinite Solutions
Jul 25, 2026Clifton CallenderSymbolic Music Generation Infinities
Florida State University
Generating in the Limit with Infinitely Many Hallucinations
Jun 8, 2026Irene Strauss, Alexandra Butoi, Ryan CotterellLarge Language Models Fail Fundamental Limits
ETH Zürich
How (and when) can you fit examples to logic-based hypothesis classes over infinite structures?
May 31, 2026Michael Benedikt, Alessio MansuttiHypothesis Class Tractability
University of Oxford, UK · IMDEA Software Institute, Spain
On Language Generation in the Limit with Bounded Memory
May 28, 2026Jon Kleinberg, Anay Mehrotra, Amin Saberi +1Bounded-Memory Learnability
Cornell University · Stanford University · Google Research
Periodic RoPE for Infinite Context LLMs
May 27, 2026Simin HuoEfficient Long-Context Inference Long-Context Reasoning
Shanghai Jiao Tong University
Optimal Non-Asymptotic Edgeworth Expansions for Multivariate Neural Network Outputs
May 22, 2026Lucia CelliNeural Approximations Large Deviation Principle
Department of Mathematics, University of Luxembourg
A Random-Matrix Criterion for Initializing Gated Recurrent Neural Networks
May 11, 2026Tommaso Fioratti, Riccardo Marcaccioli, Francesco CasolaRecurrent Neural Networks Discontinuous Gating
Capital Fund Management, 75007 Paris, France · Institute of Mathematics, EPFL, 1015 Lausanne, Switzerland
Efficient Techniques for Data Reconstruction, with Finite-Width Recovery Guarantees
May 7, 2026Edward Tansley, Roy Makhlouf, Estelle Massart +1Subspace Infinities
Mathematical Institute, Woodstock Road, University of Oxford, Oxford, UK, OX2 6GG. · ICTEAM Institute, UCLouvain, Euler Building, Avenue Georges Lemaˆıtre, 4 - bte L4.05.01, Louvain-la-Neuve, B - 1348, Belgium.
How Long Does Infinite Width Last? Signal Propagation in Long-Range Linear Recurrences
May 6, 2026Mariia SeleznovaInfinities Scaling Laws
Ludwig-Maximilians-Universität München · Max Planck Institute of Quantum Optics
Universality in Deep Neural Networks: An approach via the Lindeberg exchange principle
May 4, 2026Filippo Giovagnini, Sotirios Kotitsas, Marco RomitoInfinities Activation Functions
Topological Neural Tangent Kernel
May 1, 2026Sanjukta KrishnagopalGraph Neural Networks Neural Tangent Kernel
Department of Computer Science, University of California Santa Barbara
On the algebra of Koopman eigenfunctions and on some of their infinities
Apr 23, 2026Zahra Monfared, Saksham Malhotra, Sekiya Hajime +2Koopman Operator Spectral Decomposition
Interdisciplinary Center for Scientific Computing, University of Heidelberg, Germany. · Department of Mathematics and Computer Science, University of Heidelberg, Germany. · School of Computation, Information and Technology, Technical University of Munich, Germany & MDSI & MCML. +1
Automated Palynological Analysis System: Integrating Deep Metric Learning, Detection and Classification in Bright Field Microscopy
Apr 17, 2026J. Staforelli-Vivanco, R. Jofré, P. Coelho +12Microscopy Species Identification
Facultad de Ingeniería, Universidad Andres Bello, Talcahuano, CCP-THNO 7100 Chile · Universidad Arturo Prat, Sede Victoria, Chile · Laboratory of Palynology and Plant Ecology, School of Sciences and Technologies, University of Concepción, Los Angeles Campus
$μ$ pscaling small models: Principled warm starts and hyperparameter transfer
Feb 11, 2026Yuxin Ma, Nan Chen, Mateo Díaz +3Hyperparameter Neural Network Training
Department of Applied Mathematics and Statistics Johns Hopkins University
Computation, Condensation, and the Incompleteness Between Them: A Coupled Foundation of Intelligence
Mar 7, 2023Xin LiTractability Intelligence
Department of Computer Science, University at Albany, SUNY
A Functional-Space Mean-Field Theory of Partially-Trained Three-Layer Neural Networks
Oct 28, 2022Zhengdao Chen, Eric Vanden-Eijnden, Joan BrunaTwo-Layer Neural Networks Mean-Field Limit
Google Research Mountain View, CA 94043 · Courant Institute, New York University New York, NY 10003
Characterizing Language Generation in the Limit: Finite Witnesses and a Separation-Width Hierarchy
Date pendingXiaoyu Li, Andi Han, Jiaojiao Jiang +1Infinities Context-Free Grammars
University of New South Wales · University of Sydney

Sep 29, 2026, cs.ROJ/K move · Enter open · S save

One from Infinity: Actualizing Futures from Pretrained World Models into Robot Actions

Bang Du, Yichen Xie, Shuqi Zhao, Yuxin Chen+2

University of California, Berkeley · Southern University of Science and Technology · Xi’an Jiaotong University

Efficient World-Action Model Video World Models World Models Robocasa Infinities

Open paper AskPDF arXiv

A pretrained video world model admits many plausible futures for a scene, but a robot must realize the exact task-conditioned one. To turn world models into executable robot policies, existing methods fine-tune the heavy world model backbone using large-scale robot data and computational resources. Challenging this status quo, we argue that the expensive part has already been paid in the world model pretraining since the representation space of a video world model lays out the diverse potential futures. In this case, what remains is to select the future that accomplishes the task and to read out the actions that realize it. We formalize this task as actualization, which learns a task-conditioned selection and realization on top of a prior supplied by a frozen world model. This can be solved by a tiny actualizer model. We implement RoboActualizer with as few as 60M parameters on top of a frozen world model encoder. The actualizer is composed of two lightweight DiT experts that jointly predict future latents and actions by flow matching. The model can be trained entirely on a single GPU with 32 GB peak memory. With up to 100x fewer trainable parameters than existing WAMs and VLAs, RoboActualizer reaches great performance on simulation benchmarks including LIBERO, LIBERO-Plus, RoboTwin 2.0 and five tasks on two real-world platforms, with a low latency of 39 ms that allows real-time control.

Momentum

Related topics

Latest papers 28