Peer Revenue Play credits. Real papers.

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Tuesday 29 September

Nuclear Experiment

$\pi$-QCD; Proposal for Experiments at the GSI Pion Beam Facility

HADES Collaboration: R. Abou Yassine, J. Adamczewski-Musch, G. Appagere et al. (+126 more)

We propose to investigate the dynamics of hadronic matter in the strong QCD regime using the pion-beam facility at GSI in conjunction with HADES. This program will study hadron physics through elementary pion-nucleon interactions up to $\sqrt{s}\approx$2.35 GeV, bridging cold matter research with detailed analyses of exclusive reactions. Our goals include exploring baryon resonance formation and their electromagnetic structures. This will complement photo-production experiments (e.g., ELSA) and enhance our understanding of emissivity in dense, hot hadronic matter. Key topics include examining baryon couplings to mesons and virtual photons, analyzing hyperon weak decays in exclusive reactions, and using Partial Wave Analyses with differential cross sections and polarization observables to achieve unprecedented precision in baryon-meson coupling data ($\rho$N, $\omega$N). Furthermore, e$^{+}$e$^{-}$ production measurements off nucleons will provide insight into the electromagnetic transition form factors of baryons in the time-like region, revealing the role of vector mesons ($\rho$, $\omega$). Reactions on nuclear targets allow for detailed studies of hadron properties, including vector-meson line shapes and strengths in cold nuclear matter. These measurements provide a critical reference for the interpretation of results from the hot and dense environment created in $A+A$ collisions. This research also has broader implications. It will contribute to the modeling of neutrino-nucleus interactions, facilitate unprecedented investigations into hypernuclei formation, and enhance our understanding of strong interaction dynamics across different energy regimes.
Read the full abstract

All authors HADES Collaboration: R. Abou Yassine, J. Adamczewski-Musch, G. Appagere, M. Becker, A. Blanco, C. Blume, M. Bohman, Y. Bondar, L. Chlad, I. Ciepał, S. Deb, M. D. Doncel Monasterio, M. Duerr, W. Esmail, L. Fabbietti, M. Firlej, T. Fiutowski, A. M. Foda, J. Förtsch, P. Fonte, J. Friese, I. Fröhlich, T. Galatyuk, R. Gernhäuser, M. Grunwald, D. Grzonka, M. Gumberidze, S. Harabasz, T. Heinz, C. Höhne, F. Hojeij, R. Holzmann, M. Idzik, B. Kämpfer, K.-H. Kampert, B. Kardan, V. Kedych, S. Kim, V. Kladov, A. Kodym, M. Kohls, J. Kolas, A. Koperwas-Władyszewska, G. Korcyl, G. Kornakov, W. Krueger, A. Kugler, R. Lalik, S. Lebedev, T. Leontiou, S. Linev, F. Linz, L. Lopes, M. Lorenz, A. Malige, P. Marciniewski, J. Markert, T. Matulewicz, S. Mehat, J.G. Messchendorp, V. Metag, J. Michel, A. Molenda, J. Moron, C. Müntz, A. Mukherjee, M. Nabroth, A. Opíchal, J. Orliński, J.-H. Otto, M. Papenbrock, Y. Parpottas, M. Parschau, S. Pattnaik, C. Pauly, D. Pawlowska-Szymanska, V. Pechenov, O. Pechenova, G. Perez Andrade, J. Phan, K. Piasecki, J. Pietraszko, T. Povar, M. Prędota, K. Prościński, A. Prozorov, W. Przygoda, B. Ramstein, N. Rathod, J. T. Rieger, J. Ritman, A. Rost, A. Rustamov, S. Sahu, P. Salabura, J. Saraiva, K. Scharmann, N. Schild, K. Schönning, E. Schwab, F. Seck, I. Selyuzhenkov, J. Smyrski, M. Sobiella, S. Spies, A. Sreejith, A. Strach, H. Ströbele, J. Stroth, P. Subramani, K. Sumara, O. Svoboda, K. Swientek, J. Taylor, P. E. Tegner, P. Tlusty, M. Traxler, S. Treliński, I. C. Udrea, F. Ulrich-Pur, V. Wagner, A.A. Weber, E. Weiler, C. Wendisch, D. Wielanek, P. Wintz, H.P. Zbroszczyk, M. Zieliński, P. Zumbruch

Materials Science

Screening of Coulomb Interactions in MoS2 Nanoribbons: Enhanced Coulomb interactions, Antiscreening, and Edge Magnetism

A. Montaghemi, H. Hadipour, A. Khademi et al. (+1 more)

MoS2 has attracted significant attention for its promising applications in optoelectronics, owing to the remarkable stability of its excitons and trions. These quasiparticles have large binding energies that arise from the MoS2 moderate band gap and the unconventional screening of Coulomb interactions in low dimensions. Here, we investigate the screening of long-range Coulomb interactions in the 1H and 1T' phases of MoS2 across different dimensionalities, with particular emphasis on nanoribbons. Our analysis is based on first-principles calculations combined with the constrained random-phase approximation. This work presents a comparative study of MoS2, h-BN, black phosphorene, and graphene nanoribbons, with emphasis on the role of Coulomb interactions in shaping their electronic and magnetic properties. In one-dimensional nanoribbons, quantum confinement leads to a substantial enhancement of Coulomb interactions relative to 2D MoS2. Calculations show that the Hubbard U increases from 2.6 eV in the 2D system to nearly 3.5 eV in the semiconducting nanoribbons. In these systems, the Coulomb interaction is long-ranged, with its tail extending over more than 100 A, nearly twice of that in the 2D structure. The presence of asymmetric edge states involving d and p orbitals in non-hydrogen-passivated zigzag MoS2 nanoribbons, in contrast to phosphorene, h-BN, and graphene nanoribbons, results in a finite net magnetization.
Read the full abstract

All authors A. Montaghemi, H. Hadipour, A. Khademi, A. Yazdani

Numerical Analysis

Numerical homogenization of a linear multiscale conservation law

Claude Le Bris, Frédéric Legoll and Giulia Merlini

We address the numerical approximation of a scalar conservation law with a highly oscillatory flux, in the linear setting. Taking advantage of the intrinsic numerical viscosity of the scheme employed that yields a parabolic regularization of the original hyperbolic equation, we use homogenization theory to derive an effective equation which in turn we solve numerically. Our numerical experiments show the performance of the approach on the example of the Lax-Friedrichs scheme.
Read the full abstract

High Energy Physics (Phenomenology)

Endothermic Dark Matter at LZ from a Decaying Parent

Yongsoo Jho, Sanghwan Kim and Seong Chan Park

The LUX-ZEPLIN experiment has reported a nuclear-recoil candidate near 248 keV. We investigate a decay source for an endothermic interpretation: a long-lived dark parent produces energetic ground-state particles that upscatter on xenon. The parent mass fixes the injection energy, while its lifetime determines the absolute flux through the Galactic dark-matter column density. We include the Doppler broadening from parent and observer motion and use exact scattering kinematics with natural xenon isotopes. For a scalar contact interaction, a benchmark with a 10 MeV incident state and a 506 MeV parent gives an efficiency-weighted recoil spectrum peaked at 248.5 keV. One event in a specified high-energy window requires $\tau_X \simeq 4.4 \times 10^{20}$ s $f_X [\sigma_{\text{ref}}/(10^{-36} \text{cm}^2)]$, with a normalization sensitive to the nuclear response. We quantify the accompanying elastic-scattering requirement and the target-dependent thresholds. The calculation identifies a decay realization of the boosted flux and the portal conditions needed to preserve the recoil signal; a detector-level fit and ultraviolet-specific constraints remain necessary for a complete experimental assessment.
Read the full abstract

Artificial Intelligence

How Can Recommendation Feedback Evolve Agent Memory?

Shanwen Mao, Mingming Li, Hao Zhang et al. (+4 more)

Content-generation agents continuously receive impressions, clicks, conversions, and negative feedback from recommendation systems, providing real-world outcome signals for memory evolution. However, these signals are delayed and noisy, confounded by audience composition, placement, and recommendation policies, and may result from the combined influence of multiple memories, making accurate attribution difficult. Existing methods rely primarily on immediate feedback or semantic retrieval and therefore struggle to reliably translate recommendation outcomes into memory fitness. To address this challenge, we propose TIDE (Trajectory-Informed Directed Memory Evolution), an external memory evolution framework driven by delayed recommendation feedback. We further introduce Memory Evolution Gain (MEG), which measures the utility improvement of evolved memory over a no memory baseline on strictly future tasks. TIDE treats memory as a capacity-constrained population of experiences: temporal and semantic credit assignment estimates contextual fitness, while responsibility credit distributes outcome signals according to the memories referenced during generation. These signals are then used to reinforce, crossover, mutate, or evict memories. On an e-commerce membership marketing content-generation agent, TIDE achieves a +7.75-percentage-point MEG in offline temporal replay and significantly improves both unique click-through rate (UCTR) and activation rate in an online A/B test. On a delayed-label benchmark, TIDE achieves the lowest mean absolute error (MAE) and root mean squared error (RMSE) and the highest MEG among the compared methods, demonstrating its effectiveness.
Read the full abstract

All authors Shanwen Mao, Mingming Li, Hao Zhang, Zhiheng Li, Yige Wang, Penghua Yu, Junxiong Zhu

Computation and Language

RunyaNER: Auxiliary Language Selection for Runyankore NER

Prosper Arineitwe Asiimwe, Francois Meyer and Jan Buys

Cross-lingual zero-shot transfer and multilingual fine-tuning are promising approaches for NLP tasks such as Named Entity Recognition (NER) in low-resource languages, but in the absence of target language benchmarks, it is unclear which auxiliary language selection strategy leads to the best transfer. We introduce RunyaNER, the first publicly available NER benchmark for the East African language Runyankore, and use it to investigate the choice of which languages to use for transfer. Created with a semi-automated pipeline and fully manually verified, RunyaNER contains over 237k annotated words across 30k sentences. We benchmark pretrained models on RunyaNER, establishing that our dataset is of sufficient quality and size to produce effective Runyankore NER models. We then use RunyaNER to investigate auxiliary language selection in cross-lingual zero-shot and multilingual fine-tuning settings. Our experiments show that while transfer performance is highly sensitive to auxiliary language selection, embedding-based measures computed from labelled training spans correlate more strongly with downstream transfer performance than traditional linguistic features based on metadata or typology. By releasing RunyaNER and providing a systematic analysis of auxiliary language selection strategies, this work contributes both a new benchmark resource and practical insights for multilingual transfer in low-resource settings.
Read the full abstract

Digital Libraries

Which reported inputs govern molecular docking reproducibility? A benchmark from reporting audit to independent re-execution

Giap Duc Ha

Computational docking results can be re-executed only when the molecular system and protocol are specified, yet reporting checklists do not quantify how strongly individual method fields affect the reported score. We combined a span-verified audit of 50 open-access papers with controlled one-factor perturbations, a Vinardo scoring-function check and a Vina cross-docking extension covering 12 targets in 7 protein families, and re-execution of 37 published claims on local and independent cloud infrastructure. Search-related fields were sparsely reported, but perturbing box centre, box size, exhaustiveness or random seed produced median absolute score changes of no more than 0.08 kcal mol$^{-1}$; ligand protonation and identity produced median changes of 0.19 and 0.29 kcal mol$^{-1}$, respectively, whereas receptor structure produced the largest change (1.02 kcal mol$^{-1}$; n = 85; P < 0.001), and 22% of 116 unique reported ligand strings remained unresolved or ambiguous after deterministic name normalization. These measurements yielded empirical field weights for the NewScience Evidence score and identified exact receptor structure, machine-resolvable ligand identity and protonation state as primary reporting elements; 29 of 37 local and 27 of 37 cloud re-executions were within 2.0 kcal mol$^{-1}$ of the reported score, but these descriptive rates do not estimate experimental replication or a calibrated probability of reproduction.
Read the full abstract

Chemical Physics

The NewMag crystal-field code for f-element systems: Implementation for extended active spaces, second-order correlated energies, and generalisation to fn configurations

Gwenhaël Duplaix-Rata, Dumitru-Claudiu Sergentu, Boris Le Guennic et al. (+1 more)

We report a massive update of the NewMag program, which enables to extract Stevens crystal-field parameters (CFPs) after relativistic and multiconfigurational calculations are performed. The code can now post-treat ORCA and OpenMolcas outputs that contain state-average complete active space self-consistend field (SA-CASSCF) calculations with minimal or extended active spaces, second-order NEVPT2 or CASPT2 calculations, and spin-orbit configuration interaction (SOCI) calculations. Stevens parameters are extracted following the original Stevens convention with the so-called "extended" parameters. Rotation invariant indicators are also computed, after applying a correction of the parameter values to ensure normalization, following Rudowicz. At the SOCI level, the spin-orbit coupling (SOC) constant is also extracted within a spherical approximation of the SOC operator. For all the reference calculation levels, the model spectrum is reconstructed, allowing a direct assessment of its quality. The reported implementation is successfully applicable to f-element systems with a non-void, non-half-filled/empty or non-full f shell, that is for f^n configurations with n != 0, != 7 and != 14. Similarities and differences with the SINGLE_ANISO and AILFT codes are discussed. With selected examples, we showcase the interest of determining CFPs in f-element systems to understand their magnetic and optical properties in general, and more specifically the added value of NewMag. Finally, application of this approach to d-element systems is discussed, to reveal when it readily and successfully applies and when it may fail in reproducing satisfactorily the ab initio energies.
Read the full abstract

All authors Gwenhaël Duplaix-Rata, Dumitru-Claudiu Sergentu, Boris Le Guennic, Rémi Maurice

Sound

Rate-Agnostic Bioacoustics: Heterogeneous Multi-Taxa Classification with Continuous Filterbanks and Fourier Neural Operators

Stefano Ciapponi, Francesco Ardan Dal R{\i}, Nicola Conci et al. (+1 more)

Conventional bioacoustic classification models rely on fixed-rate spectral representations, requiring recordings acquired at heterogeneous sampling rates to be resampled before analysis. We propose a Sampling-Frequency-Independent (SFI) frontend that processes each recording directly at its native sampling rate, coupled with a Fourier Neural Operator (FNO) backbone featuring progressive temporal-scale fusion. This framework avoids fixed-rate resampling and high-frequency information loss while producing fixed-size representations across sampling rates. Mild training-time sampling-rate (sr) augmentation further improves robustness to unseen rate variations. Evaluated on a multi-taxa corpus comprising 84 classes and 60 sampling rates, the proposed SFI-FNO configuration outperforms fixed-rate and corpus-maximum-rate baselines, achieving .906 accuracy, .921 balanced accuracy, and a Macro-F1 score of .899.
Read the full abstract

All authors Stefano Ciapponi, Francesco Ardan Dal R{\i}, Nicola Conci, Elisabetta Farella

Artificial Intelligence

SkillGym: Training Skill-Use Agents with Automatic Verifiable Environment Generation

Renxi Wang, Mingshan Hee, Fajri Koto et al. (+2 more)

Skills equip LLM agents with professional knowledge and guidance to complete long-horizon and complex tasks. Although skills have been widely adopted in recent agent paradigms and harnesses, how to synthesize reliable training data and how to train agents for skill use remain underexplored. In this work, we propose SkillGym, an automatic pipeline to build verifiable environments, collect trajectories, and train skill-use agents. SkillGym first crawls a large volume of skills from the internet, then keeps those whose workflows can run reproducibly offline. A builder-reviewer pipeline is used to construct difficulty-controlled tasks, spanning four task types, each with a reference solution and an executable verifier. With this pipeline, we build 6.8k environments and collect 19k verified successful trajectories for supervised finetuning. Finetuning on these trajectories improves LLMs of different families and sizes, from 2B to 122B parameters across four skill-use benchmarks; Our Qwen3.5-9B SFT model outperforms the 397B untrained model on two of them. Further analysis shows that training teaches agents to invoke skills, raising the rate of reading the relevant skill from 28% to 96%, and that the gains hold across reasoning structures, extending to task types that form a minority of the training data and to skills held out from training
Read the full abstract

All authors Renxi Wang, Mingshan Hee, Fajri Koto, Timothy Baldwin, Haonan Li

Signal Processing

AVSG: Accelerated Vectorized Sparse Gather for Efficient KV Cache Offload in Sparse-Attention LLM Serving

Wenwei Kuang, Xiangyu Wang, Chong Wu et al. (+5 more)

Dynamic sparse attention reduces long-context attention computation by selecting only a subset of tokens, but still requires access to the full KV cache, leaving serving memory-bound. Offloading the KV cache to host memory reduces device memory pressure but places H2D transfers on the decoding critical path. In DSA, substantial overlap in selected KV entries across decoding steps creates an opportunity for device-resident reuse. Exploiting this reuse efficiently, however, presents three critical challenges: costly matching of selected tokens against entries retained in the HBM buffer, uneven distribution of the remaining H2D transfers across accelerator cores, and retaining frequently accessed KV entries within limited HBM capacity. We present AVSG, an operator that addresses these challenges for DSA while preserving its exact token selections. AVSG uses vectorized hash matching to identify reusable HBM-buffer slots and reserve slots for entries requiring transfer, then evenly partitions these H2D transfers across accelerator cores. Lifetime-based buffer management retains frequently accessed KV entries in the device buffer across decoding steps, and shared slot reservations extend reuse across tokens within a multi-token prediction iteration. On a single NPU, vectorized hash matching is 2.80x faster than scalar dual-pointer matching, miss-only transfer raises effective H2D bandwidth by up to 34.44x over request-level assignment, and an 8K-entry buffer reaches a 94.83% hit rate on real requests. These gains translate into end-to-end improvement: on a serving stack processing a real-world production dataset, AVSG reduces time per output token by 39% and increases output throughput by 1.27x relative to the same KV offload layout without HBM-buffer reuse. The code of AVSG is released at https://gitcode.com/cann/cann-recipes-infer.
Read the full abstract

All authors Wenwei Kuang, Xiangyu Wang, Chong Wu, Jun Wang, Weijie Zhang, Brian K Chen, Longwen Lan, Ken Zhang

Computer Vision

Weeding Out Bad Seeds: Initial-Noise-Robust Unlearning for Text-to-Image Diffusion Models

Arian Komaei Koma, Seyed Amir Kasaei, Aida Aryafar et al. (+6 more)

Machine unlearning has emerged as a critical post-hoc safety measure to erase sensitive concepts from Text-to-Image (T2I) models without prohibitive retraining. However, we reveal that current state-of-the-art (SOTA) approaches are brittle due to a severe lack of robustness to noise initialization. We call this phenomenon “probabilistic forgetting”: suppressed concepts re-emerge under specific random initial noise conditions, despite appearing unlearned on other initializations. We trace this failure to the misalignment between standard Gaussian sampling during unlearning and the unlearning objective. Since the target concept manifests only in specific initial noise regions throughout the unlearning phase, uniform random sampling yields sparse, uninformative gradient updates that fail to drive robust erasure. To overcome this issue, we propose an adaptive, concept-conditioned sampling strategy that dynamically concentrates gradient updates on regions where the target concept manifests, down-weighting uninformative areas. We integrate our framework with six distinct SOTA unlearning methods across four diffusion backbones and evaluate it across safety, object, and artistic-style unlearning, as well as under black-box and white-box adversarial attacks. Our method reduces the conditional nudity re-emergence rate across random initializations by 67.2% on average over four baselines and lowers attack success rates across both adversarial evaluations. Across concept domains, Adaptive Noise Sampling strengthens adversarial robustness and non-target retention while preserving competitive generative quality and target-erasure performance.
Read the full abstract

All authors Arian Komaei Koma, Seyed Amir Kasaei, Aida Aryafar, Matin Ghiasi, Ali Aghayari, Amirhossein Souri, Mohammad Mosayyebi, AmirMahdi Sadeghzadeh, Mohammad Hossein Rohban

High Energy Physics (Theory)

The Stieltjes Bootstrap for Feynman Integrals

Hadrien Brochet, Maximilian Haensch, Johannes Henn et al. (+4 more)

We prove that scalar Feynman integrals are Stieltjes functions in each kinematic variable $-p_i\cdot p_j$, as well as in the particle masses, for arbitrary loop order, arbitrary numbers of external legs, and arbitrary internal masses, whenever a certain exponent condition is satisfied. The proof rests on a canonical decomposition of the second Symanzik polynomial with manifestly non-negative coefficients indexed by spanning two-forests. We present a dedicated analysis of the case in which the kinematic variables satisfy relations, as frequently happens in physical applications. We argue that integration-by-parts relations allow the selection of master-integral bases in which every element is a Stieltjes function. The Stieltjes property guarantees the convergence of Padé approximants and provides determinantal (Hankel) constraints on the Taylor coefficients of master integrals that go beyond previously known complete-monotonicity bounds. We develop a numerical strategy that combines these Stieltjes constraints with differential equations to determine the master integrals in the Euclidean region via semidefinite optimization. We demonstrate this approach on the pedagogical example of the six-dimensional massive box integral, and then apply it to the calculation of previously unknown three-loop two-mass integrals needed for three-loop electroweak corrections to the $Z$- and Higgs-boson self-energies.
Read the full abstract

All authors Hadrien Brochet, Maximilian Haensch, Johannes Henn, Benjamin Hollering, Prashanth Raman, Qinglin Yang, Simone Zoia

Machine Learning

Why Adaptive Optimizers Underestimate Rare Tokens

Sangsidhya Kar

In the softmax output layer, a rare token receives a small positive logit gradient on most steps and a much larger negative gradient on the few steps when it is the target. SGD simply adds these contributions. Coordinate-wise adaptive methods such as Adam, RMSProp, and sign descent instead divide each update by a running estimate of its magnitude, and that estimate is largest immediately after the token appears. This imbalance has two effects. At the level of the whole output layer, we characterize which optimizers preserve the mean output embedding: every method whose update is linear in past gradients does, as do Kronecker-factored and orthogonalized methods such as Shampoo and Muon. Adam, Adafactor, Lion, and sign descent do not, and for these methods we obtain an exact step-by-step expression for the change. At the level of an individual rare token, the same normalization shifts the training fixed point. In the unigram model, sign descent lowers the logit of every token that occurs in fewer than half of the minibatches at a constant expected rate. For RMSProp with periodic arrivals, we can solve the fixed point in closed form: if a token is absent for at least two consecutive minibatches, its equilibrium probability is strictly below its data frequency for every learning rate, and the ratio tends to $\kappa/(2(e^{\kappa/2}-1))$. Here $\kappa$ is the mean number of steps between occurrences divided by the second-moment time constant $1/(1-\beta_2)$. In the same model, SGD and AMSGrad retain the unbiased fixed point. We test these predictions both in a unigram model and in a small language model trained from a known generating distribution. With random arrivals, the bias is larger than the periodic formula predicts; in the language model, the optimizers with the biased fixed point also fit the generating distribution less well.
Read the full abstract

Analysis of PDEs

Global well-posedness and scattering for energy-critical Schrodinger equations with conservative Marcus noise

Bin Liu, Fan Xu, Bo Yang et al. (+1 more)

This work establishes global well-posedness for the three-dimensional defocusing energy-critical nonlinear Schrödinger equation with conservative linear Marcus noise. For arbitrary deterministic energy-space data, we construct a probabilistically strong solution, unique in the critical spacetime class, and prove convergence of finite-activity approximations. The spatial profiles need only be bounded and Lipschitz, while the centered pure-jump Lévy noise is assumed to have finite second moment. If the temporal noise coefficient is square integrable on the half-line, the solution scatters almost surely, with uniform mean-square tail convergence and no smallness restriction. The solution conserves mass and satisfies an exact energy balance. The proof combines deterministic energy-critical stability with Poisson Strichartz estimates; convergence of the interaction-representation martingale gives the critical spacetime control needed for continuation and scattering.
Read the full abstract

All authors Bin Liu, Fan Xu, Bo Yang, Lei Zhang

Computation and Language

E-MoE: Enhanced Mixture-of-Experts for Non-Factorized Diffusion Language Models

Arseny Ivanov, Alexander Kolesov, Alexander Korotin et al. (+2 more)

Masked diffusion models (MDMs) generate sequences by progressively unmasking several tokens per denoising step, but their reverse process is typically factorized over positions, limiting sample quality in the few-step regime where diffusion's speed advantage over autoregressive decoding matters most. A recent line of work introduces a continuous Gaussian latent, trained as a variational autoencoder, to capture correlations across positions, but such approaches are prone to posterior collapse, where the latent is silently ignored. We propose Enhanced Mixture-of-Experts (E-MoE), which builds the reverse process as a mixture of factorized distributions over a discrete shared latent given by the expert-routing decisions of a Mixture-of-Experts (MoE) backbone, without increasing active parameters over the factorized baseline. Across synthetic multi-modal benchmarks, binarized MNIST, and LM1B, E-MoE improves few-step generation over factorized baselines.
Read the full abstract

All authors Arseny Ivanov, Alexander Kolesov, Alexander Korotin, Ivan Oseledets, Mikhail Goncharov

Distributed Computing

DScale: Scaling Block-Diffusion Speculative Decoding with Adaptive Verification

Rongjian Chen, Minxian Xu, Zhengxin Fang et al. (+2 more)

Growing large language model applications demand efficient inference. At high concurrency, block-diffusion speculative decoding suffers from verification padding, rejected candidates, and incompatibility between variable prefixes and fixed-shape graphs. Uniform truncation sacrifices acceptable tokens. We present DScale, preserving drafter architecture, weights, and full draft length. A separate 112K-parameter predictor requires neither confidence calibration nor hardware speed-curve preparation. Path-aware tiles reduce padding. Dynamic verify-length (DVL) allocation packs scored prefixes into half the native verification capacity. Fixed-address workspaces propagate changing boundaries through verification and acceptance while reusing captured graphs. On A100-40GB with tensor parallelism 1, Qwen3-8B and Qwen3-4B cover four datasets and concurrency 8-32, reusing each target's frozen predictor. Geometric-mean throughput gains across these configurations are respectively 43.9% and 48.8% over DFlash, 22.2% and 37.7% over DSpark, and 24.4% and 32.0% over Domino, with lower request latency. Cumulative ablations show that adding the three mechanisms successively increases geometric-mean throughput, while budget adjustment improves accepted-token retention. GPU profiling shows that complete decode-step time on GSM8K decreases by 30.8-52.5% relative to DFlash
Read the full abstract

All authors Rongjian Chen, Minxian Xu, Zhengxin Fang, Kejiang Ye, Chengzhong Xu

Software Engineering

Profiling the Energy Consumption of Serverless Functions with Joule Profiler

Jérémy Woirhaye, François Gibier, Anderson Andrei Da Silva et al. (+3 more)

Cloud providers and customers have widely adopted serverless computing as a convenient paradigm for deploying and executing functions on demand. To do so, serverless platforms require provisioning an appropriate execution environment before a single line of the function's code runs. These environments consist of several layers, such as container engines, hypervisors, unikernels, and programming language runtimes. While the literature has investigated the performance of these serverless platforms, it treats functions as black boxes, and the community lacks key insights into the environmental impacts of packaging applications as serverless functions. This paper therefore empirically studies the energy efficiency of serverless functions deployable on serverless platforms. We design an experimental benchmarking environment that lets stakeholders explore the impacts of the various layers involved in executing serverless functions. We use it to evaluate 1,401 configurations, combining 9 execution environments, 7 language-runtime configurations, 11 workloads, and 3 input sizes, to answer three research questions: Are the most popular programming languages for serverless functions the most energy-efficient? What factors most affect their energy efficiency? What are the most energy-efficient configurations to deploy them? Our results show that one should first choose the programming language, then the language runtime, and only then the execution environment, which matters only for short-lived functions and whose best choice depends on the runtime. Our benchmarking environment, experimental artifacts, raw measurements, and analysis code are publicly available.
Read the full abstract

All authors Jérémy Woirhaye, François Gibier, Anderson Andrei Da Silva, Francescomaria Faticanti, Thomas Ledoux, Romain Rouvoy

Robotics

RawVLA: Embodied Neural Image Signal Processor For Robotic Manipulation

Shuhong Liu, Heng Zhou, Lingfeng Qian et al. (+7 more)

Vision-language-action (VLA) models typically operate on RGB images produced by a fixed camera image signal processor (ISP), leaving the imaging pipeline outside the learning and evaluation loop. We systematically examine the consequences of this overlooked design choice across five fundamental ISP dimensions: gain, sensor noise, chromatic response, tonal response, and bit depth. Our analysis reveals that RAW-to-RGB processing materially shapes both action prediction and manipulation success, with different ISP dimensions exerting substantially different effects. Guided by these findings, we introduce RawVLA, a streaming neural ISP that adaptively renders RAW observations for frozen VLA policies while concentrating its capacity on the imaging factors relevant to embodied behavior. We further present RawVLA-Bench, a RAW-domain manipulation benchmark to expose image processing as an explicit evaluation variable across clean and adverse acquisition conditions. Experiments on RawVLA-Bench show that RawVLA preserves performance under standard conditions while substantially improving robustness under degraded imaging, establishing adaptive RAW processing as an effective interface between physical cameras and embodied policies.
Read the full abstract

All authors Shuhong Liu, Heng Zhou, Lingfeng Qian, Yuhao Fang, Xianbao Hou, Qianyu Zhou, Lin Gu, Wei Sui, Jianfei Yang, Ziteng Cui

Computer Vision

Principled MAP estimation for inverse problems: bridging the gap between convergence and performance

Alexandre Lagier, Valentine Tosel, Anne Gagneux et al. (+2 more)

Pretrained denoisers provide a powerful way to incorporate image priors into restoration algorithms. Plug-and-Play and RED approaches exploit fixed-noise-level denoisers within first-order optimization schemes, with convergence guarantees, but often struggle to achieve high-quality reconstruction on severely ill-posed inverse problems. In contrast, recent state-of-the-art approaches leverage denoisers derived from flow- or diffusion-based generative models and evaluate them along a sequence of decreasing noise levels. While these methods achieve strong empirical performance, their convergence theory remains limited. In this paper, we bridge this gap by specifically designing an algorithm that combines denoisers at decreasing noise levels with a schedule tailored to ensure convergence. From a Bayesian perspective, we prove that our method converges to a $\textit{Maximum a Posteriori}$ (MAP) estimate, under suitable assumptions. Subsequently, we apply our method to various ill-posed inverse problems and show that it surpasses convergent methods while competing with state-of-the-art empirical ones.
Read the full abstract

All authors Alexandre Lagier, Valentine Tosel, Anne Gagneux, Mathurin Massias, Ségolène Martin

Optimization and Control

A Spatial Benders Algorithm for Delay Constrained Routing

Antonio Frangioni, Laura Galli, Luca Mencarelli et al. (+1 more)

We study the Delay Constrained Routing (DCR) problem arising in IP computer networks to support high bandwidth applications. The goal is to route IP packets, from source to destination, subject to quality of service constraints – in particular, bounding the worst case delay – while minimizing the total allocated bandwidth in the network. DCR can be formulated as a Mixed-Integer Second-Order Cone Program, and solved using standard solvers; while generally efficient, this hinges on off-the-shelf tools and does not scale to large instances. We devise a solver-free bespoke approach that uses nested Benders-Lagrange, namely, a double decomposition approach that exploits the special structure of the problem. In particular, we use a nonconvex nonlinear formulation of the problem, that results in a nonconvex Benders value function whose optimization we tackle with a bespoke Spatial Branch-and-Bound approach using custom piecewise-convex cuts. We show that our approach provides very tight upper and lower bounds in competitive running times w.r.t. previous state-of-the-art approaches.
Read the full abstract

All authors Antonio Frangioni, Laura Galli, Luca Mencarelli, Enrico Sorbera

Quantum Physics

Digital quantum state preparation by nucleation

Jean Paul Sadia, Morten Hjorth-Jensen, Dean Lee et al. (+1 more)

Quantum state preparation is crucial for quantum algorithms, and improved state preparation methods can reduce resource requirements by orders of magnitude. In this work, we introduce a state preparation method for lattice Hamiltonians on digital quantum computers inspired by nucleation. In our method, the exact ground state of a small lattice Hamiltonian is prepared in a quantum circuit, then a series of growth stages are performed via Trotterized adiabatic evolution to increase the size of the lattice to the desired model. Using the adiabatic theorem and Trotter error bounds, we provide sufficient conditions on the total evolution time and number of Trotter steps required for nucleation to prepare an initial state with desired fidelity. In addition, we introduce optimized nucleation which considers variational optimization on top of the nucleation structure. For a two-dimensional Ising model, we show that optimized nucleation can outperform a current leading method for state preparation on digital quantum computers.
Read the full abstract

All authors Jean Paul Sadia, Morten Hjorth-Jensen, Dean Lee, Ryan LaRose

Software Engineering

Exploring Emotional Intelligence in Software Testing

Cristina Martinez Montes, Pushpamalar Rajendran, Grace Karuri et al. (+1 more)

Background: Emotional Intelligence (EI) is the ability to recognise, understand, and manage one's own and others' emotions. Software testers deliver judgements about colleagues' work under deadlines they do not control, and prior work on emotion in software engineering has mostly studied developers. Aims: To explore how software testers describe the part EI plays in their day-to-day work, in communication and conflict within the team, and in responding to requirements volatility. Method: Semi-structured interviews with 16 software testers in Sweden working in teams that use agile practices, across aviation, automotive, healthcare, IT services, administration, banking and pharmaceuticals, analysed with reflexive thematic analysis informed by Goleman's EI framework. Results: Three themes. Testers described regulating stress under deadline pressure and drawing motivation from recognition, clarity and autonomy; managing the daily delivery of critical findings to colleagues so that trust survives; and responding to requirements change with frustration that turned into decisions about what to leave untested, into advocacy for process change, or into workarounds. Read against developer-focused studies, the themes point to features of the testing role: the work product is a criticism of a colleague's work, success is invisible while failure is attributed, and the tester's window shrinks with every upstream delay. Conclusions: For testers, managing emotions is a constant job requirement. The results highlight that the importance of EI increases when the development process lacks an independent testing phase. The findings also inform implications for teams and, ultimately, for organisations and future research.
Read the full abstract

All authors Cristina Martinez Montes, Pushpamalar Rajendran, Grace Karuri, Felix Dobslaw

Machine Learning

Physical Muon: Orthogonalization as an Equilibrium Computation

Yuren Hao

Physical neural networks and analog in-memory computing could reduce the energy cost of neural network training. Realizing this potential, however, requires optimizers that combine effective learning with physical implementability. SGD fits local analog updates but struggles on transformers, while Adam family is unstable against analog bias. Muon offers strong training performance, but its Newton–Schulz orthogonalization relies on dense matrix-matrix products. To address this obstacle, we introduce Physical Muon, which computes the orthogonalization as the equilibrium of a continuous-time flow. Random probes approximate the flow using matrix-vector products, reciprocal reads, and local rank-1 writes. To test whether this replacement preserves training performance, we evaluate it on a 10.95M-parameter transformer. The dense flow's mean validation cross-entropy is 0.0085 above Newton–Schulz across nine seeds per method; the probe implementation is 0.0188 above the control across two seeds. Circuit simulations further reproduce the flow dynamics and yield comparable training behavior.
Read the full abstract

Quantum Algebra

Structure and Representations of a Diagrammatic Non-Commutative Tangloid Algebra

Hadeel B. Albeladi and Sofia Lambropoulou

We introduce the tangloid algebra $\mathcal{T}_n$, a diagrammatic algebra arising from the extended tangloid category $\Ti$, and its subalgebra, the braidoid algebra $\mathcal{B}_n$. The construction begins with the unoriented tangloid category $UTC$, inspired by Turaev's theory of knotoids, which is obtained from the unoriented welded tangleoid category $UWTC$ by removing the welded relation while retaining the forbidden moves and imposing some additional relations. Within this framework, the braidoid category arises naturally as a subcategory of $UTC$. We then introduce the extended tangloid category $\Ti$ by adjoining a placeholder morphism $\emptyset$, which allows the generators to be extended to endomorphisms of a fixed $n$-box. The interaction of the placeholder morphism with the crossing morphisms gives rise to diagonal morphisms, $/$ and $\backslash$, which are used in the defining relations of $\Ti$. The corresponding extended braidoid category $\mathbf{Brd}_{\Ti}$ is realised as a subcategory of $\Ti$. For each $n\geq 0$, we then define the tangloid algebra $\mathcal{T}_n$ by linearising the endomorphism algebra $\operatorname{End}_{\Ti}(n)$, as a unital $\mathbb{C}$–algebra presented by diagrammatic generators and relations. We also define the braidoid subalgebra $\mathcal{B}_n$, which provides an algebraic framework for the theory of braidoids. We further introduce the reduced tangloid algebra $\delta\mathcal{T}_n$ by imposing two additional reduction relations for trivial knot and trivial knotoid components, and we construct a natural diagrammatic bilinear pairing on $\delta\mathcal{T}_n$ using diagrammatic reflection, composition, and closure. Our categorical and algebraic setting provides a theoretical framework for studying diagrammatic structures with intrinsic endpoints, namely knotoids, linkoids, and braidoids, and related diagram algebras.
Read the full abstract

Quantum Physics

Asymptotic Pseudospectra in Dissipative Floquet Quantum Systems: Geometric Structures and Observable Dynamics

Yuncheng Xie, Haozhe Shi, Zhuocheng Ma et al. (+2 more)

In periodically driven open quantum systems, nonnormality renders the Floquet spectrum insufficient as the system approaches the thermodynamic limit, so that pseudospectra are needed to characterize the dynamics accurately. While conventional approaches mainly focus on the local dynamics of isolated pseudospectra, their global connections and the resulting physical consequences for observables have remained largely unexplored. Here, we uncover this collective behavior by classifying the unit disk into distinct domains of exponential, algebraic, and bounded accuracy according to the asymptotic size-scaling laws of pseudospectral residuals, yielding an underlying geometric structure. We demonstrate the physical implications of this structure through two exactly solvable models. First, in a dissipative shift chain, parameter tuning drives geometric transitions that are detectable via spin-wave observables. Second, in a chiral XY model, we use this geometric structure to explain the origin of a measurable phenomenon: two correlation signals exchange their retention order under continuous parameter tuning. Our findings not only establish a new theoretical paradigm for understanding dissipative Floquet quantum systems, but also predict geometry-driven observables for quantum computing experiments.
Read the full abstract

All authors Yuncheng Xie, Haozhe Shi, Zhuocheng Ma, Weibin Chu, Xin-Gao Gong

Machine Learning

Graph-Conditioned On-Policy Agent Distillation from Off-the-Shelf Teachers

Xiaohan Yi, Wen Luo, Yani Huang et al. (+4 more)

On-policy distillation (OPD) trains compact language agents with teacher feedback on student-generated trajectories. In multi-turn tasks, compounding errors can move students beyond the teacher's effective supervision. We introduce Graph-Conditioned On-Policy Agent Distillation (GC-OPD), which enriches an off-the-shelf teacher's scoring context with execution evidence. A graph indexes repeated teacher executions by shared states while preserving complete successful and failed histories. After each student episode, GC-OPD retrieves current-state references or historical alternatives and combines them with student hindsight to score the original thought-action tokens. Using the same original teachers, GC-OPD improves mean success over vanilla OPD from 24.70% to 48.78% on ScienceWorld (4B student), from 53.36% to 85.26% on ALFWorld Unseen, and from 29.10% to 37.65% on WebShop. At matched student sizes, it also achieves higher mean success than every evaluated OPD baseline using GRPO-trained teachers on ScienceWorld and ALFWorld; the strongest such ScienceWorld 4B baseline reaches 46.66%. GC-OPD requires no task-specific teacher optimization.
Read the full abstract

All authors Xiaohan Yi, Wen Luo, Yani Huang, Junfeng Zhan, Asher Qin, Peilin Zhao, Xi Xiao

Physics and Society

Cultural, Structural, and Mediated Routes to Polarization in a Nonlinear Mean-Field Model

Jérôme Michaud and Fredrik Jansson

Collective polarization can arise through mechanisms acting at different stages of social communication, but similar polarized outcomes need not imply dynamically equivalent processes. We introduce a nonlinear two-group mean-field model that distinguishes three routes to polarization: individual content filtering, assortativity, and receiver-side message tailoring. Sender-side reformulation is included as a complementary mediation process that transforms source signals before they are mixed. These message transformations are motivated by AI-mediated cultural transmission, in which intermediaries can reformulate messages or adapt them to receivers. Using common and difference coordinates, we derive general conditions for instabilities toward consensus and balanced polarization and analyze the resulting stationary branches in the full two-dimensional phase plane. Individual filtering can generate polarization through state-dependent updating, assortativity can preserve group differences in the social field, and receiver tailoring can create a polarizing feedback even under complete mixing. By contrast, sender-side reformulation affects the polarization mode only when assortativity preserves differentiated source signals. Although these mechanisms can produce identical local instability conditions or balanced-branch amplitudes, they remain distinguishable through transverse stability, asymmetric equilibria, multistability, and basin geometry. In particular, balanced polarization may emerge as a saddle before becoming a stable attractor, while asymmetric saddles organize the boundary between consensus and polarization basins. These results show that collective polarization depends not only on feedback strength, but also on where feedback enters the communication process.
Read the full abstract

Category Theory

Six functor formalisms via internal higher algebra

Shachar Carmeli, Guy Kapon and Noam Nissan

We extend a six-functor formalism $D\colon\mathrm{Span}(C,E)\to\mathrm{Cat}$ to a lax symmetric monoidal functor of $(\infty,2)$-categories $\mathbf{Span}^2(C,E)^P_I\to\mathbf{Cat}$, where $P$ and $I$ are the classes of $D$-proper and $D$-étale morphisms, respectively. This proves a conjecture of Mann and generalizes a special case of a theorem of Cnossen, Lenz, and Linskens. To prove this result, we develop a theory of internal $\mathsf{E}$-monoidal categories and $\mathsf{E}$-operads, where $\mathsf{E}$ is a local class of morphisms in an $\infty$-topos. These notions generalize the internal symmetric monoidal categories and operads developed by Martini and Wolf.
Read the full abstract

Robotics

Video2STL: Grounding VLM-Generated Temporal Specifications for Robot Learning

Merve Atasever, Keyan Azbijari, Cagan Bakirci et al. (+6 more)

Video-based policy learning is particularly promising, as it illustrates target behaviors without requiring action annotations or embodiment-matched demonstrations. A central challenge is deciding what information should be transferred from the video to the robot. Existing approaches commonly convert visual observations into scalar similarity or value signals, or ask foundation models to directly generate reward code. These approaches can make the temporal structure of a task difficult to inspect, ground, and reuse. We present Video2STL, a framework that converts observation-only videos into parametric Signal Temporal Logic (STL) specifications and uses the resulting formal representation for robot learning. A vision-language model extracts an embodiment-independent semantic event trace and constructs a bank of symbolic temporal specifications. The model determines the task structure, while numerical predicate thresholds and temporal bounds are grounded from successful robot trajectories. For policy learning, we separate short- and long-timescale temporal information: short-horizon specifications provide dense rewards through rolling-window quantitative robustness, while a causal monitor over a retained long-horizon specification provides one-time progress rewards for valid temporal prefixes. The same representation supports cross-embodiment transfer from human or animal videos to robot control. Across four manipulation tasks, Video2STL achieves $85.8\%$ average success-once and $67.0\%$ success-at-end, compared with $81.5\%/59.5\%$ for native dense PPO and $65.0\%/42.3\%$ for Text2Reward; in quadruped locomotion, Qwen-3.8 and GPT-5.6-based Video2STL policies achieve $100\%$ success across velocities from $0.3$ to $2.1\,\mathrm{m/s}$ while remaining competitive in high-speed energy efficiency. Project webpage: \href{https://video2stl.github.io/}{video2stl}.
Read the full abstract

All authors Merve Atasever, Keyan Azbijari, Cagan Bakirci, Bo-Ruei Huang, Tolga Izdas, Zahra Shahrooei, Richard Yang, Erdem Biyik, Jyotirmoy V. Deshmukh