Tootfinder

OMGSR: You Only Need One Mid-timestep Guidance for Real-World Image Super-Resolution
Zhiqiang Wu, Zhaomang Sun, Tong Zhou, Bingtao Fu, Ji Cong, Yitong Dong, Huaqi Zhang, Xuan Tang, Mingsong Chen, Xian Wei
https://arxiv.org/abs/2508.08227

OMGSR: You Only Need One Mid-timestep Guidance for Real-World Image Super-Resolution
Denoising Diffusion Probabilistic Models (DDPM) and Flow Matching (FM) generative models show promising potential for one-step Real-World Image Super-Resolution (Real-ISR). Recent one-step Real-ISR models typically inject a Low-Quality (LQ) image latent distribution at the initial timestep. However, a fundamental gap exists between the LQ image latent distribution and the Gaussian noisy latent distribution, limiting the effective utilization of generative priors. We observe that the noisy laten…

Another timestep of the #Pleiades #occultation by the #Moon, at 21:26 UTC - Merope had just emerged behind the dark side. Interestingly no stars whatsoever could be seen in 11 x 70 binoculars though the Moon - nice terminator - was a beauty. The Moon's elevation was 15°; 1/4 second at f/3.2 and ISO 200.

Streaming Sequence-to-Sequence Learning with Delayed Streams Modeling
Neil Zeghidour, Eugene Kharitonov, Manu Orsini, V\'aclav Volhejn, Gabriel de Marmiesse, Edouard Grave, Patrick P\'erez, Laurent Mazar\'e, Alexandre D\'efossez
https://arxiv.org/abs/2509.08753

Streaming Sequence-to-Sequence Learning with Delayed Streams Modeling
We introduce Delayed Streams Modeling (DSM), a flexible formulation for streaming, multimodal sequence-to-sequence learning. Sequence-to-sequence generation is often cast in an offline manner, where the model consumes the complete input sequence before generating the first output timestep. Alternatively, streaming sequence-to-sequence rely on learning a policy for choosing when to advance on the input stream, or write to the output stream. DSM instead models already time-aligned streams with a …

And the the #Pleiades #occultation by the #Moon is already history in Bochum, Germany: a final timestep at 22:34 UTC, with the Moon having cleared all major stars of the cluster. Nice show that was - though only for the camera (which was a Panasonic DMC-FZ300 on a tripod, mostly at maximum zoom; this image 1/5 second at f/3.2 and ISO 200, with the Pleiades 25° up).

On the convergence of N-body simulations of the Solar System
Hanno Rein, Garett Brown, Mei Kanda
https://arxiv.org/abs/2507.04987 https://

On the convergence of N-body simulations of the Solar System
Most direct N-body integrations of planetary systems use a symplectic integrator with a fixed timestep. A large timestep is desirable because simulations run fast. However, simulations yield unphysical results if the timestep is too large. Surprisingly, no systematic convergence study has been performed on long (Gyr) timescales. In this paper we present numerical experiments to determine the minimum timestep one has to use in long-term integrations of the Solar System in order to recover the sy…

IML-Spikeformer: Input-aware Multi-Level Spiking Transformer for Speech Processing
Zeyang Song, Shimin Zhang, Yuhong Chou, Jibin Wu, Haizhou Li
https://arxiv.org/abs/2507.07396

IML-Spikeformer: Input-aware Multi-Level Spiking Transformer for Speech Processing
Spiking Neural Networks (SNNs), inspired by biological neural mechanisms, represent a promising neuromorphic computing paradigm that offers energy-efficient alternatives to traditional Artificial Neural Networks (ANNs). Despite proven effectiveness, SNN architectures have struggled to achieve competitive performance on large-scale speech processing task. Two key challenges hinder progress: (1) the high computational overhead during training caused by multi-timestep spike firing, and (2) the abs…

Reinforcement Learning with Action Chunking
Qiyang Li, Zhiyuan Zhou, Sergey Levine
https://arxiv.org/abs/2507.07969 https://arxiv.org/pdf/2507.07969 https://arxiv.org/html/2507.07969
arXiv:2507.07969v1 Announce Type: new
Abstract: We present Q-chunking, a simple yet effective recipe for improving reinforcement learning (RL) algorithms for long-horizon, sparse-reward tasks. Our recipe is designed for the offline-to-online RL setting, where the goal is to leverage an offline prior dataset to maximize the sample-efficiency of online learning. Effective exploration and sample-efficient learning remain central challenges in this setting, as it is not obvious how the offline data should be utilized to acquire a good exploratory policy. Our key insight is that action chunking, a technique popularized in imitation learning where sequences of future actions are predicted rather than a single action at each timestep, can be applied to temporal difference (TD)-based RL methods to mitigate the exploration challenge. Q-chunking adopts action chunking by directly running RL in a 'chunked' action space, enabling the agent to (1) leverage temporally consistent behaviors from offline data for more effective online exploration and (2) use unbiased $n$-step backups for more stable and efficient TD learning. Our experimental results demonstrate that Q-chunking exhibits strong offline performance and online sample efficiency, outperforming prior best offline-to-online methods on a range of long-horizon, sparse-reward manipulation tasks.
toXiv_bot_toot

Timestep-Compressed Attack on Spiking Neural Networks through Timestep-Level Backpropagation
Donghwa Kang, Doohyun Kim, Sang-Ki Ko, Jinkyu Lee, Hyeongboo Baek, Brent ByungHoon Kang
https://arxiv.org/abs/2508.13812

Timestep-Compressed Attack on Spiking Neural Networks through Timestep-Level Backpropagation
State-of-the-art (SOTA) gradient-based adversarial attacks on spiking neural networks (SNNs), which largely rely on extending FGSM and PGD frameworks, face a critical limitation: substantial attack latency from multi-timestep processing, rendering them infeasible for practical real-time applications. This inefficiency stems from their design as direct extensions of ANN paradigms, which fail to exploit key SNN properties. In this paper, we propose the timestep-compressed attack (TCA), a novel fr…

BTPG-max: Achieving Local Maximal Bidirectional Pairs for Bidirectional Temporal Plan Graphs
Yifan Su, Rishi Veerapaneni, Jiaoyang Li
https://arxiv.org/abs/2508.04849 https://…

BTPG-max: Achieving Local Maximal Bidirectional Pairs for Bidirectional Temporal Plan Graphs
Multi-Agent Path Finding (MAPF) requires computing collision-free paths for multiple agents in shared environment. Most MAPF planners assume that each agent reaches a specific location at a specific timestep, but this is infeasible to directly follow on real systems where delays often occur. To address collisions caused by agents deviating due to delays, the Temporal Plan Graph (TPG) was proposed, which converts a MAPF time dependent solution into a time independent set of inter-agent dependenc…

TLB-VFI: Temporal-Aware Latent Brownian Bridge Diffusion for Video Frame Interpolation
Zonglin Lyu, Chen Chen
https://arxiv.org/abs/2507.04984 https://

TLB-VFI: Temporal-Aware Latent Brownian Bridge Diffusion for Video Frame Interpolation
Video Frame Interpolation (VFI) aims to predict the intermediate frame $I_n$ (we use n to denote time in videos to avoid notation overload with the timestep $t$ in diffusion models) based on two consecutive neighboring frames $I_0$ and $I_1$. Recent approaches apply diffusion models (both image-based and video-based) in this task and achieve strong performance. However, image-based diffusion models are unable to extract temporal information and are relatively inefficient compared to non-diffusi…

SDSNN: A Single-Timestep Spiking Neural Network with Self-Dropping Neuron and Bayesian Optimization
Changqing Xu, Buxuan Song, Yi Liu, Xinfang Liao, Wenbin Zheng, Yintang Yang
https://arxiv.org/abs/2508.10913

SDSNN: A Single-Timestep Spiking Neural Network with Self-Dropping Neuron and Bayesian Optimization
Spiking Neural Networks (SNNs), as an emerging biologically inspired computational model, demonstrate significant energy efficiency advantages due to their event-driven information processing mechanism. Compared to traditional Artificial Neural Networks (ANNs), SNNs transmit information through discrete spike signals, which substantially reduces computational energy consumption through their sparse encoding approach. However, the multi-timestep computation model significantly increases inferenc…

Time-Aware One Step Diffusion Network for Real-World Image Super-Resolution
Tainyi Zhang, Zheng-Peng Duan, Peng-Tao Jiang, Bo Li, Ming-Ming Cheng, Chun-Le Guo, Chongyi Li
https://arxiv.org/abs/2508.16557

Time-Aware One Step Diffusion Network for Real-World Image Super-Resolution
Diffusion-based real-world image super-resolution (Real-ISR) methods have demonstrated impressive performance. To achieve efficient Real-ISR, many works employ Variational Score Distillation (VSD) to distill pre-trained stable-diffusion (SD) model for one-step SR with a fixed timestep. However, due to the different noise injection timesteps, the SD will perform different generative priors. Therefore, a fixed timestep is difficult for these methods to fully leverage the generative priors in SD, …

Diff-TONE: Timestep Optimization for iNstrument Editing in Text-to-Music Diffusion Models
Teysir Baoueb, Xiaoyu Bie, Xi Wang, Ga\"el Richard
https://arxiv.org/abs/2506.15530 …

Diff-TONE: Timestep Optimization for iNstrument Editing in Text-to-Music Diffusion Models
Breakthroughs in text-to-music generation models are transforming the creative landscape, equipping musicians with innovative tools for composition and experimentation like never before. However, controlling the generation process to achieve a specific desired outcome remains a significant challenge. Even a minor change in the text prompt, combined with the same random seed, can drastically alter the generated piece. In this paper, we explore the application of existing text-to-music diffusion …

Sparse, Geometry- and Material-Aware Bases for Multilevel Elastodynamic Simulation
Ty Trusty, David I. W. Levin, Danny M. Kaufman
https://arxiv.org/abs/2508.13386 https://

Sparse, Geometry- and Material-Aware Bases for Multilevel Elastodynamic Simulation
We present a multi-level elastodynamics timestep solver for accelerating incremental potential contact (IPC) simulations. Our method retains the robustness of gold standard IPC in the face of intricate geometry, complex heterogeneous material distributions and high resolution input data without sacrificing visual fidelity (per-timestep relative displacement error of $\approx1\%$). The success of our method is enabled by a novel, sparse, geometry- and material-aware basis construction method whi…

Neural Field Turing Machine: A Differentiable Spatial Computer
Akash Malhotra, Nac\'era Seghouani
https://arxiv.org/abs/2509.03370 https://arxiv.org/pd…

Neural Field Turing Machine: A Differentiable Spatial Computer
We introduce the Neural Field Turing Machine (NFTM), a differentiable architecture that unifies symbolic computation, physical simulation, and perceptual inference within continuous spatial fields. NFTM combines a neural controller, continuous memory field, and movable read/write heads that perform local updates. At each timestep, the controller reads local patches, computes updates via learned rules, and writes them back while updating head positions. This design achieves linear O(N) scaling t…

IMEX-RB: a self-adaptive IMEX time integration scheme exploiting the RB method
Micol Bassanini, Simone Deparis, Francesco Sala, Riccardo Tenderini
https://arxiv.org/abs/2506.16470

IMEX-RB: a self-adaptive IMEX time integration scheme exploiting the RB method
In this work, we introduce a self-adaptive implicit-explicit (IMEX) time integration scheme, named IMEX-RB, for the numerical integration of systems of ordinary differential equations (ODEs), arising from spatial discretizations of partial differential equations (PDEs) by finite difference methods. Leveraging the Reduced Basis (RB) method, at each timestep we project the high-fidelity problem onto a suitable low-dimensional subspace and integrate its dynamics implicitly. Following the IMEX para…

On Fixed-Parameter Tractability of Weighted 0-1 Timed Matching Problem on Temporal Graphs
Rinku Kumar, Bodhisatwa Mazumdar, Subhrangsu Mandal
https://arxiv.org/abs/2508.10562 ht…

On Fixed-Parameter Tractability of Weighted 0-1 Timed Matching Problem on Temporal Graphs
Temporal graphs are introduced to model systems where the relationships among the entities of the system evolve over time. In this paper, we consider the temporal graphs where the edge set changes with time and all the changes are known a priori. The underlying graph of a temporal graph is a static graph consisting of all the vertices and edges that exist for at least one timestep in the temporal graph. The concept of 0-1 timed matching in temporal graphs was introduced by Mandal and Gupta [DAM…

Quantization Meets Spikes: Lossless Conversion in the First Timestep via Polarity Multi-Spike Mapping
Hangming Zhang, Zheng Li, Qiang Yu
https://arxiv.org/abs/2508.14520 https:/…

Quantization Meets Spikes: Lossless Conversion in the First Timestep via Polarity Multi-Spike Mapping
Spiking neural networks (SNNs) offer advantages in computational efficiency via event-driven computing, compared to traditional artificial neural networks (ANNs). While direct training methods tackle the challenge of non-differentiable activation mechanisms in SNNs, they often suffer from high computational and energy costs during training. As a result, ANN-to-SNN conversion approach still remains a valuable and practical alternative. These conversion-based methods aim to leverage the discrete …

Explicit Monotone Stable Super-Time-Stepping Methods for Finite Time Singularities
Zheng Tan, Tariq D. Aslam, Andrea L. Bertozzi
https://arxiv.org/abs/2507.17062 https://…

Explicit Monotone Stable Super-Time-Stepping Methods for Finite Time Singularities
We explore a novel way to numerically resolve the scaling behavior of finite-time singularities in solutions of nonlinear parabolic PDEs. The Runge--Kutta--Legendre (RKL) and Runge--Kutta--Gegenbauer (RKG) super-time-stepping methods were originally developed for nonlinear complex physics problems with diffusion. These are multi-stage single step second-order, forward-in-time methods with no implicit solves. The advantage is that the timestep size for stability scales with stage number $s$ as $…