Peer Revenue Equity in science

Papers

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

All of arXiv

Saturday 29 August

Machine Learning

PathBridger: Subgoal Bridges for Offline Goal-Conditioned Reinforcement Learning

Offline goal-conditioned reinforcement learning (GCRL) aims to learn policies for reaching diverse goals entirely from fixed trajectory data. Long-horizon offline GCRL remains challenging because sparse goal-reaching signals must be propagated over many steps, while execution errors cannot be corrected through additional environment interaction. Existing methods address these challenges by improving long-range value estimation or reducing the effective decision horizon through subgoals, options, and action chunks. In several hierarchical methods, however, a selected subgoal specifies where to go, while the intervening state-space path remains implicit in an endpoint-conditioned low-level policy. To address this interface, we propose PathBridger, a hierarchical offline GCRL method that explicitly connects subgoal selection to short-horizon execution. PathBridger constructs a state-space bridge toward the selected intermediate endpoint and decodes it into a short executable action chunk using an inverse dynamics model. Experiments across the evaluated OGBench tasks demonstrate strong aggregate performance, with particularly large gains on the multi-object Cube manipulation tasks. Code: https://github.com/SChoish/PathBridger
Read the full abstract

Computational Geometry

Bellman–Shoreline Search in Arbitrary Dimension: Exponential Vector Oscillators, Active Memory, Precession, and Effective Computability

We study online search for an unknown affine hyperplane in $\mathbb{R}^D$, for arbitrary fixed finite dimension. Building on a companion self-similar cell reduction and support-function formulation, we ask how the mechanism changes as the normal space grows from $\mathbb{S}^0$ to $\mathbb{S}^{D-1}$. In $D=1$, alternation and productivity yield an equal-ripple principle and the exact stationary constant $9$. In $D=2$, the analogous relative equilibrium is a logarithmic spiral whose bottleneck chord imposes tangency and selects the pitch. For exponential orbits $\Gamma(\sigma)=e^{\kappa\sigma}\omega(\sigma)$, we develop log-directional geometry, exponentially discounted memory, gauges, and recursive hyperspherical parametrizations. Without a shape ansatz, the bottleneck admits a certificate supported by at most $D$ historical suppliers, and at globally worst phases the current point lies on the active face. Within regular chambers we derive exact variation, tangency, pitch, age, and, in $D=3$, delay-system identities. Odd-dimensional obstructions, antipodal subclasses, and harmonic towers provide constraints and explicit candidate families but are not claimed globally optimal. Finally, the N-COMP theorem shows that $C_D^*$ is a computable real for every fixed finite $D$ and that algebraic polygonal $\varepsilon$-optimal cells can in principle be synthesized. Numerical screening through $D=10$ is kept separate from the proved results.
Read the full abstract

Cosmology and Nongalactic Astrophysics

Primordial black holes and induced gravitational waves from localized features in DBI inflation

We investigate primordial black hole formation in a Dirac-Born-Infeld inflationary framework in which the background dynamics are determined by explicit functional forms of the inflaton potential $V(\phi)$ and the warp factor $f(\phi)$. The background equations are solved numerically to obtain the evolution of the slow-roll parameters. We show that a localized feature in the potential, accompanied by a correlated structure in the warp factor, dynamically induces a transient suppression of the slow-roll parameter, leading to a short-lived non-attractor phase. This behavior generates an enhancement of the curvature power spectrum on small scales, while preserving consistency with CMB-scale observables. The Mukhanov-Sasaki equation is then solved numerically to compute the resulting power spectrum, which exhibits narrow and localized peaks in the PBH abundance across different mass ranges. We also evaluate the associated stochastic background of induced gravitational waves and find that the predicted signal can fall within the sensitivity bands of future pulsar timing arrays and space-based interferometers, depending on the scale of the inflationary feature. A parameter scan in the $(r,n_s)$ plane shows that the large-scale predictions remain consistent with current observational constraints.
Read the full abstract

Computation and Language

RouteSparse: Input-Conditional Pattern Routing for Budgeted Long-Context Prefilling

Dynamic sparse attention can reduce the quadratic cost of long-context prefilling without changing model weights. MInference assigns each attention head one pattern offline and estimates that pattern's sparse indices for every prompt. This design is efficient, but it assumes that a head's preferred pattern and sparsity budget remain suitable across inputs. We introduce RouteSparse, which routes each head and prompt segment among a small library of GPU-efficient sparse patterns. A low-cost probe estimates pattern utility and uncertainty; a latency-aware router then selects a pattern and budget, while uncertain cases fall back to a denser mask. We formulate routing as constrained risk minimization, derive an attention-output error certificate from omitted probability mass, and evaluate the method on long-context retrieval, question answering, summarization, and language modeling. On Llama 3.1-8B-Instruct with 128K-token prompts, RouteSparse achieves $6.5\times$ dense prefill speed with a 0.2-point RULER drop relative to dense attention, compared with $7.3\times$ speed and a 1.6-point drop for fixed per-head routing. Ablations confirm that input-conditional routing, hardware profiling, and selective dense fallback each contribute to the quality–latency tradeoff.
Read the full abstract

All authors Chao Zhang, Yifan Ji, Ziyan Zhang, Kai Song, Fei Lin

Machine Learning

Sparse Koopman Autoencoders Identify Local Dynamical Regimes in Multibasin Systems

Koopman autoencoders (KAEs) seek a higher-dimensional latent representation in which nonlinear dynamics evolve linearly. However, many interesting systems have multiple basins of attraction, and both theoretical and empirical work has shown these multibasin systems cannot generally admit a single finite-dimensional global Koopman embedding under standard assumptions. We posit that encoders with a sparsity-inducing objective encouraging few active latent coefficients will provide latent supports as an inspectable basin-modeling principle for Koopman autoencoders. We use these encoders producing sparse latents in training Sparse Koopman Autoencoders (SKAEs) without basin labels or other regime annotations, and treat the learned latent supports as model-produced regime variables after training. Across a range of procedurally generated multibasin systems and chaotic flows, we show that SKAEs have superior forecasting performance compared to dense-latent KAEs. We also perform a mechanistic study that shows latent supports produced by SKAEs are both essential for the quality of the representation and useful for identifying basins on basin-interior states, whereas dense-latent KAEs collapse to an uninformative single support family. These results identify sparse latents and their corresponding supports as label-free, interpretable regime variables for Koopman learning in nonlinear systems with multiple local dynamical laws.
Read the full abstract

All authors Aidan Li, Uday Kiran Reddy Tadipatri, Mahan Fathi, Sarath Chandar, Ross Goroshin

Information Theory

Neural Network-Based Delay-Doppler-Assisted Channel Estimation for OFDM

Conventional orthogonal frequency division multiplexing (OFDM) channel estimation relies on single-tap estimation and time-frequency (TF) interpolation, which becomes unreliable in high-mobility channels because Doppler-induced inter-carrier interference (ICI) invalidates the underlying element-wise TF model. This paper proposes a neural-network-based delay-Doppler (DD)-assisted channel estimation framework for OFDM over doubly selective channels. We first derive an ICI-aware TF domain input-output relation and formulate channel estimation as a DD recovery problem. Unlike conventional sparse recovery approaches, the proposed framework does not require the equivalent DD domain channel vector to be strictly sparse, thereby accommodating the leakage induced by fractional delay and Doppler shifts. Since the data symbols are unknown during channel estimation, the sensing matrix is constructed using only the known pilot symbols. As a result, data-induced interference is not explicitly modeled, leading to a structured mismatch in the pilot observations. To tackle this challenge, the adopted network iteratively exchanges observation- and channel-domain features through the sensing matrix to learn the mapping from these contaminated observations to the equivalent DD domain channel, which is subsequently used to reconstruct the TF-domain channel. Simulation results show that the proposed method achieves lower normalized mean-square error and bit-error rate than conventional OFDM estimators.
Read the full abstract

All authors Mingcheng Nie, Hao Chang, Shuangyang Li, Haiyao Yu, Jiafu Hao, Yonghui Li

Computers and Society

Why Organizational Rules Fail AI: O-I-B-A-R and the Externalization of Decision Boundaries

AI systems increasingly enter organizations through policies, procedures, playbooks, prompts, and other explicit representations of work. Yet formal descriptions often differ from situated practice, and captured know-what can omit the contextual know-how experts use when judgments are uncertain. We argue that a recurring class of organizational AI failures arises partly from a knowledge representation problem at the sociotechnical interface: the AI receives the procedure, while the organization operates on the procedure plus negative boundaries, runtime judgments, responsibility assignments, and learning history. We introduce O-I-B-A-R (OPEN, IS, BUT, ACTION, RESULT), a scaffold for externalizing these missing decision boundaries. IS records when a judgment holds. BUT records a concrete failure containing information beyond the logical negation of IS. Comparable success and failure cases are decomposed toward a minimally sufficient changing variable, which becomes a value-bearing decision dimension. A suspension represents the state in which the dimension is known but its current value is unresolved, specifying what must be measured, asked, retrieved, or escalated to a human. RESULT confirms a boundary, shifts a threshold, or exposes a new dimension. Incidents can generate new dimensions, unresolved values can define human-AI handoffs, and feedback can expand the decision space. We also identify a sociotechnical tension: durable and attributable failure histories can suppress the candor on which useful boundary knowledge depends. Externalization must therefore be designed as an organizational intervention with real costs and incentives.
Read the full abstract

Artificial Intelligence

Let Prompts Bridge Defense Knowledge: Transferable Graph Purification via Vulnerability-Aware GPL

Graph Neural Networks (GNNs) have emerged as a cornerstone for representing complex relational dependencies in diverse multimedia tasks, particularly in cross-platform user interest modeling and cross-modal semantic alignment. In the real world, a practical defense against graph adversarial perturbations is needed. However, we observe that the prevailing adversarial purification methods are essentially domain-restricted defenses, which leads to the following shortcomings: (1) single-domain data provides insufficient structural and semantic diversity for learning robust purification criteria; (2) training of domain-specific defense strategies from scratch consumes substantial computational cost. To address the above limitations, we propose a transferable graph purification scheme, named ProGAP, to bridge adversarial defense knowledge via vulnerability-aware graph prompt learning. Firstly, to capture universal adversarial patterns, a perturbation-capture edge detector is pretrained on data-rich graphs by jointly modeling topological and semantic information. Subsequently, to achieve more knowledge transfer w.r.t. robustness, vulnerability-aware prompts are designed that inject targeted purification guidance into biased nodes, during which the pretrained detector adapts to distribution shifts in downstream graphs without parameter-laborious updates. Experimental results demonstrate that compared with state-of-the-art baselines, our ProGAP achieves 1%-9% improvement, and reduces the time consumption by up to 2.2x. The code for ProGAP is available at https://github.com/Lieyoufffff/ProGAP.
Read the full abstract

All authors Shuomin Xue, Jingyuan Li, Ju Jia, Jingxuan Yu, Xiaojun Jia

Distributed Computing

Performance Evaluation of RED-ONION: A High-Speed Disk-to-Disk Transfer System

Modern experimental instruments produce data faster than general-purpose file transfer interfaces can move it, so delivery to the computing infrastructure has become a bottleneck in the research process. At many universities and research institutes, moreover, the instruments that generate research data and the high-performance computing systems that analyze it are separated both geographically and organizationally, because each demands its own expertise and installation environment. Connecting the two seamlessly is a pressing challenge for data-driven science. This article presents RED-ONION, a high-speed disk-to-disk transfer system that connects research facilities, on campus and beyond, to a computing center. The system combines data transfer nodes, a dedicated high-bandwidth network, an all-flash parallel file system, and multi-threaded transfer software that parallelizes network transmission and storage access. The design targets the wire rate both along the entire path, from the read on the sender storage to the write on the receiver storage, and for a single file between one pair of nodes rather than only in aggregate over many files or nodes. We describe the end-to-end optimizations across the transfer software, the operating system, and the storage that this requires. We evaluate a prototype deployed over a 100 Gbps transpacific path between Atlanta and Tokyo with a 150 ms round-trip time, on which a single 1 TB file transfer reached 90 Gbps, delivering a terabyte in approximately 95 s. Moving a dataset of this size therefore becomes a routine step, and the computing center serves an instrument as if the two were co-located.
Read the full abstract

All authors Keichi Takahashi, Hiroaki Kataoka, Takeo Hosomi, Ayahiro Takaki, Yasunori Kakizawa, Shuichi Ihara, Nobuaki Hashizume, Susumu Date

Information Theory

DeepHSIC: Deep Learning-based Signal Detector for Hybrid Downlink IM-NOMA

DeepHSIC is introduced as a neural receiver for hybrid downlink IM-NOMA transmission. The considered scheme combines power-domain NOMA with a composite OFDM/OFDM-IM waveform, so that user information is mapped jointly onto constellation symbols, subcarrier-index patterns, and different power levels. Although maximum-likelihood detection can achieve strong reliability for this model, its search space grows rapidly with the number of users and subcarriers. Conventional SIC reduces part of this burden, but its sequential cancellation may still accumulate errors and does not fully exploit the structure of IM-NOMA signals. To address this limitation, the proposed detector embeds dedicated deep neural network modules into the receiver and replaces the most computationally demanding SIC operations with learned inference blocks. The receiver is trained for Rayleigh fading channels and uses preprocessed channel-output features to recover user symbols. Simulation results show that DeepHSIC reaches BER performance close to model-based detectors under both perfect and imperfect CSI while requiring substantially lower detection time. These results indicate that learned SIC-style detection is a practical candidate for scalable hybrid downlink IM-NOMA receivers.
Read the full abstract

All authors Dung Nguyen Tran, Toan D. Gian, Tien-Hoa Nguyen, Mai Xuan Trang, Tien-Cuong Nguyen, Thien Van Luong

High Energy Physics (Phenomenology)

Taming lepton portal dark matter by a non-invertible selection rule

We point out that a non-invertible selection rule can control the flavor violation inherent in lepton portal dark matter, a weakly interacting massive particle candidate that can evade severe constraints from direct detection experiments. We show that ordinary invertible symmetries cannot suppress flavor-violating portal couplings while accommodating the large mixing angles in the PMNS matrix. We then present a simple construction based on the $\mathbb{Z}_2$ gauging of a $\mathbb{Z}_5$ symmetry, which realizes one-flavor dominance of the portal couplings consistently with the PMNS mixing matrix. We also discuss possible flavor leakage in a type-I seesaw completion and study experimental probes of the model through lepton observables, especially lepton flavor non-universality in $Z$ boson decays.
Read the full abstract

Instrumentation and Methods for Astrophysics

First overnight balloon flight of the GRAINE 2023 emulsion gamma-ray telescope enabled by a large-scale pressure-vessel gondola

The Gamma-Ray Astro Imager with Nuclear Emulsion (GRAINE) project conducts precision observations of sub-GeV–GeV cosmic gamma rays using a balloon-borne nuclear-emulsion telescope with high angular resolution. In GRAINE 2023, a 2.5-m$^{2}$ telescope was flown in the project's first overnight balloon flight, including observation periods for the Vela pulsar and Galactic center region. To operate the telescope under the low-pressure and low-temperature stratospheric environment, the balloon-style pressure-vessel concept was scaled up to a lightweight gondola with an internal length of 4.9 m. A new aluminum-alloy ring structure and a lightweight membranous-shell material, SHL-300MDL, were developed. While the telescope aperture was increased by a factor of 6.6 over GRAINE 2018, the pressure-vessel gondola mass was limited to 179 kg. Ground tests of the completed flight assembly demonstrated a differential pressure above 100 hPa at room temperature and at a mean temperature of $-66.0^{\circ}$C. The payload was launched from Alice Springs, Australia, in April 2023 and achieved a total flight duration of approximately 27 h, including 24.3 h of level flight. Although the upper membranous shell reached approximately $-60^{\circ}$C at night, the vessel internal pressure remained above the required 100 hPa throughout level flight. These results demonstrate that the developed gondola can accommodate a 2.5-m$^{2}$ emulsion gamma-ray telescope and maintain the required pressure during overnight stratospheric flight. Scientific analyses of astrophysical and atmospheric gamma rays, including dedicated analysis of the Galactic center region, are ongoing using the recovered emulsion data. This development provides a technical basis for repeated observations with future large-area GRAINE telescopes.
Read the full abstract

All authors Hiroki Rokujo, Shigeki Aoki, Takashi Azuma, Hirotaka Hayashi, Yudai Isayama, Atsushi Iyono, Takumi Kato, Tsuyoshi Kawahara, Koichi Kodama, Ryosuke Komatani, Masahiro Komatsu, Masahiro Komiyama, Hideyuki Minami, Kunihiro Morishima, Fumiya Murakami, Shogo Nagahara, Naotaka Naganawa, Mitsuhiro Nakamura, Tomoaki Nakamura, Yuya Nakamura, Noboru Nakano, Toshiyuki Nakano, Kazuma Nakazawa, Miyuki Oda, Kazuhiro Okamoto, Osamu Sato, Kai Shimizu, Amon Suganami, Yuki Sugi, Kou Sugimura, Satoru Takahashi, Ikuya Usuda, Saya Yamamoto, Jun Yamashita, Mayu Yamashita, Shoma Yoneno, Masahiro Yoshimoto

Quantum Physics

Eigenvalues of multipartite entanglement witnesses

We investigate various properties of multipartite block-positive operators, decomposable EWs (DEWs), and non-decomposable EWs. We provide a necessary and sufficient condition to construct a special DEW using the multipartite GHZ state. For multipartite DEW, we explicitly characterize the supremum and infimum of maximum and minimum eigenvalues, as well as the trace of its square. We also derive other results concerning NDEW, eigenvalues of $2 \times n$ EWs and the corresponding physical implications. Furthermore, we investigate the tightness of inequalities of these eigenvalues with examples.
Read the full abstract

Computer Vision

DReSG: Diffusion Residuals for Stylized Gaussian Splatting

Reference-guided stylization of scenes represented by 3D Gaussian Splatting (3DGS) is important for efficient and controllable 3D content creation. Existing VGG-feature-based 3D stylization methods provide stable rendered-view optimization, but often under-represent expressive reference style cues; diffusion models offer stronger image priors, yet direct per-view or score-based diffusion guidance can lead to view drift, local artifacts, and hard-to-control appearance updates. We present DReSG, a 3D-grounded residual-feedback framework for stylized Gaussian splatting. DReSG represents attention-guided diffusion proposals as residual targets relative to the current render, and progressively absorbs these residuals into a shared Gaussian scene through multi-view Gaussian feedback. To make this feedback stable and controllable, DReSG modulates residual strength during target construction and combines coverage-aware view selection with conflict-filtered color updates during multi-view fitting. Extensive experiments demonstrate that DReSG achieves competitive reference-guided stylization while better preserving scene structure and cross-view stability. Our project page is available at https://vpx-ecnu.github.io/DReSG-website/.
Read the full abstract

Complex Variables

Class VII surfaces with $b_2=3$ and two foliations are Kato

Let $S$ be a minimal compact complex surface of class VII with $b_2(S)=3$. We prove that if $S$ carries two distinct singular holomorphic foliations, then $S$ contains a global spherical shell. Moreover, it is an Inoue-Hirzebruch surface. The proof is based on Teleman's theorem on the existence of a cycle of rational curves and results of Dloussky and is inspired by work of Brunella.
Read the full abstract

Geometric Topology

Fox $p$-Colorings as Fixed Points of Braid Representations

Fox $p$-colorings of knots and links admit a linear-algebraic description in terms of braid representations. For each braid $\beta \in B_n$ we associate a representation $M : B_n \to \mathrm{GL}(n,\mathbb{F}_p)$ such that Fox $p$-colorings of the closed braid $\widehat{\beta}$ correspond to fixed points of $M(\beta)$. As a consequence, $$ \mathrm{Col}_p(\widehat{\beta}) = p^{\dim \ker(M(\beta)-I)}, $$ yielding an explicit formula for the number of Fox $p$-colorings of arbitrary knots and links. We apply the method to torus links and spiral links, obtaining explicit descriptions of their Fox 3-coloring spaces.
Read the full abstract

All authors Neungjoo Goh, Suah Jeong, Seungbin Oh, Hyungkee Yoo

Machine Learning

NVE: A Separability and Coverage-Aware Internal Validation Metric for Biclustering

Biclustering, or co-clustering, aims to discover coherent submatrices by grouping rows and columns of a data matrix simultaneously. This local two-dimensional structure makes validation more difficult than in ordinary clustering, where internal indices usually rely on compactness and separation in a single shared feature space. Existing popular internal biclustering measures such as Mean Squared Residue (MSR), and Virtual Error (VE) mainly evaluate within-bicluster coherence. Although useful, these measures do not directly assess whether the extracted biclusters are mutually distinct or whether they explain a meaningful portion of the data matrix. This paper investigates Normalised Virtual Error (NVE), an internal validation metric that extends VE using a super-bicluster normalization strategy. By comparing the VE of each bicluster with the VE obtained after merging it with other biclusters, NVE introduces a relative notion of separability and redundancy. We also study a coverage-adjusted variant, NVE\textsubscript{cov}, which penalizes solutions that obtain low error by selecting only very small submatrices. Through controlled synthetic benchmarks and yeast gene-expression datasets, we examine whether NVE and NVE\textsubscript{cov} provide information beyond standard coherence-based metrics. The results show that NVE is sensitive to redundant and poorly separated biclusters, while NVE\textsubscript{cov} changes solution rankings when low-error biclusters cover only a negligible part of the matrix. These findings suggest that NVE-based measures are useful complementary criteria for internal co-clustering validation, especially when coherence, separability, and coverage must be considered jointly.
Read the full abstract

All authors Paritosh Tiwari, I Navin Kumar, James C. Bezdek, Punit Rathore

Computer Vision

Di$^2$CycleSB: Towards High-Quality Unsupervised Nighttime Visibility Enhancement via Schrödinger Bridge Transformer

Light-effect contamination poses a significant challenge to nighttime visibility enhancement. Most methods suppress light effects by estimating and decomposing them through prior-driven regularization, yet they are often limited by hand-crafted priors and ill-posed nature of decomposition. This work proposes Di$^2$CycleSB, a unsupervised Cycle Schrödinger Bridge Transformer framework guided by dynamic integral image priors, for high-quality unsupervised nighttime visibility enhancement. Specifically, a novel light-effect estimator is introduced to parameterize Gaussian-like adaptive priors by aggregating dynamic integral image representations for non-uniform glow estimation. Then, we propose a prior-informed Generator that exploits light-effect representations to guide long-range dependency modeling within our specific Transformer blocks. We formulate light-effect suppression as a Schrödinger bridge problem and construct forward and backward bridges with cycle consistency constraints to achieve visually pleasing enhancement. Extensive experiments on real-world datasets demonstrate the remarkable effectiveness of our Di$^2$CycleSB in enhancing nighttime visibility. In particular, it achieves effective end-to-end light-effect suppression without any regularization constraints and image decomposition. The code and models are available at https://github.com/LHTcode/Di2CycleSB.
Read the full abstract

All authors Hanting Li, Xin Sun, Wei Ye, Jungong Han, Liang-jie Zhang

High Energy Physics (Phenomenology)

Towards power corrections in the factorization of baryon quasi-distribution amplitudes in LaMET

Light-cone distribution amplitudes (LCDAs) are essential to precision phenomenological studies. They can be accessed from lattice QCD through the large-momentum effective theory (LaMET) via quasi-distribution amplitudes (quasi-DAs). Factorization of quasi-DAs receive power corrections in inverse powers of the hadron momentum, including target-mass and higher-twist corrections. In this work, we present the first systematic analysis of such power corrections for the leading-twist baryon quasi-DA. Establishing the moment relation between the quasi-DA and the LCDA, we derive an exact closed-form relation that resums target-mass correction to all orders at leading twist. This result also applies to heavy baryons and to quasi-transverse-momentum-dependent distributions. We numerically assess these corrections for the $\Lambda$ baryon quasi-DA using existing lattice data, finding that the target-mass correction decreases rapidly with increasing baryon momentum and is almost negligible in the endpoint regions. In addition, we explicitly construct the next-to-leading-twist operators entering the quasi-DA factorization. Our results are a first step toward quantifying the power corrections in future lattice determinations of light or heavy baryon LCDAs.
Read the full abstract

Chemical Physics

Simulating spectrally resolved exciton–exciton-interaction pump–probe spectroscopy

Exciton–exciton-interaction pump–probe spectroscopy (EEI–PP) captures the fifth-order signal of light-matter interaction, directly probing exciton–exciton annihilation. This allows studying excitation energy transport even in extended, nearly isoenergetic landscapes. Although spectrally resolved fifth-order signals have already been explored in simplified cases, in larger systems only spectrally integrated signals have been analyzed. We present a microscopic theoretical description for spectrally resolved EEI–PP response with realistic lineshapes, revealing how single- and multi-exciton state structure shape the spectral features and dynamics. The description is applicable to large molecular systems such as squaraine polymers, facilitating a direct comparison with experimental spectra. By a decomposition of the total signal into individual excitation pathways, we demonstrate that the spectral resolution is needed to distinguish otherwise intertwined single- and multi-exciton dynamics, such as single-exciton relaxation and exciton–exciton annihilation.
Read the full abstract

Human-Computer Interaction

Sharing Roughness with Hand-Outline Visualization to Reduce Sensory Asymmetry in VR Collaboration

In collaborative VR, asymmetric access to haptic hardware creates a critical information gap: tactile evidence remains private to the haptic user, hindering the shared understanding needed for joint decision-making. While prior work has explored crossmodal sensory cues in virtual environments, it remains unclear how such cues should be designed for asymmetric collaboration, where collaborators receive information through different modalities. In our setting, the haptic user feels roughness through fingertip vibration, whereas the non-haptic user relies on vision alone. To reduce this asymmetry, we propose externalizing an object's tactile state through a glanceable hand-outline visual proxy. Specifically, we examine whether abstract visual roughness cues based on line shape and motion can encode three discrete roughness levels for both haptic and non-haptic users. Two preliminary studies establish a shared visual semantics by identifying visually distinguishable cues for non-haptic users and validating their visuo-haptic correspondence for haptic users. In a main study of a collaborative sorting task, showing this visualization on both users' hands significantly reduced completion time relative to a no-visualization baseline. Moreover, NU-side cue visibility was associated with higher confidence and perceived contribution for the non-haptic user. These findings show that hand-anchored abstract visual cues provide a lightweight means of externalizing object tactile state, reducing information asymmetry without compromising social presence.
Read the full abstract

All authors Minju Baeck, Yoonseok Shin, Hyunjin Lee, Boram Yoon, Sang Ho Yoon, Woontack Woo

Commutative Algebra

Weakly Newton-nondegenerate binomial ideals

An ideal $I$ of a polynomial ring $R=k[x_1,\dots,x_n]$ is called weakly Newton-nondegenerate, or weakly NND, if its integral closure $\overline{I}$ is a monomial ideal. We study weak Newton nondegeneracy for the family of quadratic binomial ideals \[ I=(x_1^2+\epsilon_1x_{a_1}x_{b_1},\ \dots,\ x_n^2+\epsilon_nx_{a_n}x_{b_n}),\qquad \epsilon_i\in\{\pm1\},\ a_i\neq b_i, \] over an algebraically closed field. We prove that $I$ is weakly NND if and only if $\overline I=\mathfrak{m}^2$, if and only if the given generators form a regular sequence, and if and only if an explicit combinatorial condition on the pair (support pattern, sign pattern) holds: no nonempty subset $S\subseteq\{1,\dots,n\}$ is simultaneously closed for the support data and sign-trivial for the associated lattice of relations. The last equivalence rests on a solvability criterion for systems of monomial equations over a divisible abelian group, in the spirit of Eisenbud and Sturmfels. As an application, we classify all such ideals for $n=3$.
Read the full abstract

Computer Vision

sRGB Real Noise Modeling via Noise-Aware Sampling with Normalizing Flows

Noise poses a widespread challenge in signal processing, particularly when it comes to denoising images. Although convolutional neural networks (CNNs) have exhibited remarkable success in this field, they are predicated upon the belief that noise follows established distributions, which restricts their practicality when dealing with real-world noise. To overcome this limitation, several efforts have been taken to collect noisy image datasets from the real world. Generative methods, employing techniques such as generative adversarial networks (GANs) and normalizing flows (NFs), have emerged as a solution for generating realistic noisy images. Recent works model noise using camera metadata, however requiring metadata even for sampling phase. In contrast, in this work, we aim to estimate the underlying camera settings, enabling us to improve noise modeling and generate diverse noise distributions. To this end, we introduce a new NF framework that allows us to both classify noise based on camera settings and generate various noisy images. Through experimental results, our model demonstrates exceptional noise quality and leads in denoising performance on benchmark datasets.
Read the full abstract

All authors Dongjin Kim, Donggoo Jung, Sungyong Baik, Tae Hyun Kim

Computer Vision

DocIntent: Answerability-Guided Agentic Restoration for Real-World Document Visual Question Answering

Real-world degradations such as blur, shadow, distortion, and moire patterns severely impair the document question-answering capabilities of Multimodal Large Language Models (MLLMs). Applying restoration tools before Visual Question Answering (VQA) is an intuitive solution. However, existing restoration approaches remain limited, as manually designing and executing restoration strategies is labor-intensive and requires domain expertise. Agentic restoration offers new possibilities for automation, yet existing frameworks primarily target natural images and pursue perceptual quality, overlooking that restoration should serve downstream tasks rather than optimize generic image quality metrics. To this end, we explore the value of agentic restoration for real-world degraded document VQA and propose DocIntent, a training-free Answerability-Guided Agentic Restoration framework. DocIntent first assesses question answerability, then identifies task-relevant degradations and selectively invokes restoration tools. A Comparison-Based Rollback mechanism validates each restoration step and reverts it when question-relevant evidence becomes less decipherable. The entire process requires no additional pretrained degradation classifier or image quality assessment model. Extensive experiments on the WildDoc benchmark show that DocIntent consistently improves the average score and consistency of different open- and closed-source MLLMs. The code and experimental data will be publicly available.
Read the full abstract

All authors Zihan Huang, Shihang Wu, Junle Liu, Peirong Zhang, Yongxin Shi, Xuhan Zheng, Lianwen Jin

Networking

SmartBAN on Silicon by Structured Behavioral Modeling

Wireless body area networks (WBANs) underpin the Internet of Medical Things (IoMT). IEC/ETSI SmartBAN defines a lightweight time-division multiple access (TDMA) protocol, yet published implementations on commercial hardware remain partial. The standard specifies frame formats but leaves device-internal behavior in phase control, connection lifecycle, and slot timing unspecified. We present an operational SmartBAN realization on a commercial wireless system-on-chip (Nordic nRF54L15, Arm Cortex-M33, Zephyr RTOS) that implements initial connection, scheduled uplink, and timeout-based disconnection detection through a four-layer software architecture whose hardware abstraction layer is the intended portability boundary. Hub and Node Mealy automata supply one concrete design for the unspecified phase control and connection lifecycle, as a blueprint for the firmware's event-driven dispatch. Validation with 16 concurrent Nodes over 25 hours shows conformance across every transition the build exercises (all but one), >99.99% packet delivery, and timeout-triggered reconnection after the single observed disconnection. An analytical current model fits the measurements once it uses measured state currents; with datasheet currents it underestimates the average, mostly in the radio-off state, whose current we attribute to implementation-level overhead rather than the hardware. A same-SoC comparison with Bluetooth Low Energy (BLE) shows how each protocol departs from its schedule: SmartBAN, configured to send a fresh sample every interval instead of retransmitting, kept each Node's frame spacing within 1.1 ms of the interval but lost some frames; BLE lost none but deferred some by one or two intervals, and drew less than half SmartBAN's average current.
Read the full abstract

All authors Masato Yoshimi, Takahiro Ito, Kento Tanaka, Hirokazu Tanaka

Artificial Intelligence

EmoLASP: Emotion Recognition with Language Models and Answer Set Programming

Emotion recognition in conversations is increasingly tackled with language models, but these models can be unstable and expensive to fine-tune or to prompt with long dialogue histories. We propose EmoLASP, a framework that combines a language model with declarative reasoning via Answer Set Programming (ASP) to predict VAD scores (Valence-Arousal-Dominance) in conversations. Experiments on a widely used benchmark dataset (IEMOCAP) across six open-source LLMs (3B-120B) and two PLMs (BERT, RoBERTa) show that EmoLASP improves prediction performance compared to using the language model alone, even when the LLMs/PLMs are given no dialogue history in their prompts or input vectors. The gains are largest for prompt-only LLMs, which EmoLASP uses without any fine-tuning. However, for fine-tuned PLMs, the reasoner adds little once dialogue history is available. EmoLASP's LLM pipeline demonstrates the potential advantages of using a reasoning approach to ensure emotion prediction consistency and to reduce both the cost of fine-tuning and the cost of prompting with long dialogue histories.
Read the full abstract

Computation and Language

A Unifying Perspective on Language Model Representations: From Filler-Role Structure to Mechanistic Interpretability

A wide range of methods have been proposed for interpreting language models, delivering important insights into their inner workings. However, different methods and their resulting insights stand in relative isolation: what could the underlying structure of language models be, such that they give rise to all our interpretations? In this work, we propose using Tensor Product Representations (TPRs) as a unifying hypothesis. TPRs give a concrete proposal for how compositional structure could be represented in vector space — as filler-role bindings. We show, both mathematically and empirically, that TPRs can unify several prior interpretability methods: additive analogies, linear probing, sparse autoencoders, and activation patching. Mathematically, we show that these methods can all be derived from TPRs. Empirically, we apply the derivations to a range of different models — from small toy models to LLMs — to construct instances of each of the above interpretability methods; these constructed variants perform comparably to their standard variants. We view this work as a step toward what interpretability will ideally provide: a unified account of the nature of neural networks, corroborated not just by individual observations but also by an explanation of the connections between them.
Read the full abstract

Fluid Dynamics

Physics-Guided Machine Learning for Predictive Turbulent Combustion Simulation

Predictive simulation of turbulent combustion remains challenging because stiff chemical kinetics, turbulent transport, molecular mixing, and heat release interact nonlinearly across unresolved scales. In the filtered or averaged formulations, reaction rate is unclosed, and the integration of detailed chemistry constitutes the dominant computational cost. Conventional closures such as mixing-limited models, flamelet and manifold methods, conditional moment closure, and transported probability/filtered density function (PDF/FDF) formulations, encode substantial physical insight, but rely on structural assumptions that may lose validity outside their calibrated regimes. Machine learning offers a complementary approach because of its ability to approximate complex nonlinear functions. In turbulent combustion, this capability can be used to relax restrictive closure assumptions, learn unresolved nonlinear mappings from data, and accelerate expensive computations such as detailed chemistry computation. However, purely data-driven models may violate conservation laws and thermochemical consistency. They may also extrapolate poorly outside the training domain and destabilize the CFD solvers in which they are embedded. Physics-guided machine learning (PGML) addresses these failure modes by incorporating prior knowledge throughout the modeling pipeline: in the training data and input features, the model architecture, the loss function, the hybrid closure structure, and the solver-aware validation protocol. This article reviews PGML for turbulence-chemistry interaction (TCI) modeling within this framework.
Read the full abstract

Computational Geometry

Bellman Search in Arbitrary Finite Dimension: A Self-Similar Cell Theorem and Effective Computability of Planar Shoreline Search

A shoreline-search path starts at the origin and must meet an unknown affine line, without knowing either its normal or its distance. We first establish a self-similar reduction theorem for homogeneous search problems whose historical information is a record profile updated by pointwise maximum. Two quasi-returns of the normalized state delimit a block that renews the required profile by itself; a short connector closes this block into a cell. Every finite-ratio path can therefore be approximated, with arbitrarily small loss, by repetitions of a single cell at all scales. The main chain is then made effective. A finite coding of the state space computably bounds the scale factor and normalized length of a nearly optimal cell. For planar Shoreline search, the support function of the convex hull gives an exact cell functional. A one-sided polygonalization then reduces the problem to a computable number of vertices, after which quantifier elimination decides whether a polygonal cell exists below a rational threshold. It follows that the optimal deterministic planar Shoreline value $C_2^*$ is a computable real: for every rational $\epsilon>0$, an algorithm terminates with a rational interval of width at most $\epsilon$ containing $C_2^*$. Additional results—sliding memory, Bellman transitions, deadlines, geometric filters, and relative equilibria—are presented separately as a toolbox for certified computation and for the study of spiral rigidity; they are not used in the computability proof.
Read the full abstract