Peer Revenue Publish or profit

Papers

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

All of arXiv

Friday 4 September

Machine Learning

How to Speculate about Uncertainty in Agentic Coding? A Draft-Model Gate Method

LLM agents deployed for software engineering fail expensively: they act confidently wrong, and bad actions are recognized only after costly execution and retry. We present Speculative Uncertainty (SU), a method that recovers a predictive failure signal for a black-box agent from its output tokens alone, with no access to logits, weights, activations, or repeated sampling. Inverting speculative decoding, a small open-weight draft model scores the agent's already-generated trajectory in a single forward pass. From these speculative cross-likelihoods we extract phase-aware features by separating the reasoning and action spans, and calibrate them against a verifiable objective. SU produces a failure-likelihood score that any downstream policy, such as routing, human intervention, or extra test-time compute, can consume directly. To show the signal is actionable, we instantiate one such policy, a pre-execution veto gate, on software engineering agents Qwen3-Coder-480B and closed-source Claude 3.5 Sonnet, cutting execution error rate by 6-8 percentage points and token cost by 14-19% in deployment, transferring to out-of-distribution benchmarks without retraining, and generalizing across agent models.
Read the full abstract

Applied Physics

Thermoelectric properties of metal-semiconductor percolation composites

We analyze thermoelectric response of composite materials comprising homogeneous semiconductor hosts with randomly distributed metallic inclusions. We demonstrate that the properties of such materials are strongly affected by the percolation transition (with background conductivity). We show that in vicinity of percolation threshold, the power factor and the thermoelectric figure of merit strongly depend on both the concentration and arrangement of inclusions. Finally, we identify promising regimes for enhancing the power factor in realistic composites.
Read the full abstract

All authors Oleksandr Porubaniyi, Sergii Tomko, Mihailo Cherkez, Viktor Podolskiy, Radion Cherkez

Solar and Stellar Astrophysics

ROCKETS I: Investigating the Impact of the Rocket Effect on Two Nearby Young Open Clusters Encased in Infrared Bubbles

We introduce ROCKETS (Rocket-driven Cluster Kinematics & Triggered Star-formation), a Gaia based programme investigating kinematic fingerprints of stellar feedback or rocket effect in young star forming regions. We analyse two young clusters with O-type stars enclosed by mid-infrared bubbles, Collinder 69 and IC 1396 as a methodological demonstration on two nearby benchmark regions. Co-moving candidates are selected within an astrometric ellipse defined by the median and median absolute deviation of Class I/II YSOs with Gaia data. For each star, we compute a relative proper-motion angle (RPMA) relative to the ionising source. In both regions the RPMA distribution shows a strong excess at RPMA $<15^{\circ}$, corresponding to sources moving away from the respective ionising sources. We quantify this using the Rocket Effect Index (REI), finding $\mathrm{REI}\simeq 1.1$ for Collinder 69 and $\mathrm{REI}\simeq 1.7$ for IC 1396. When considering only Class I/II YSOs, the REI of both regions increases to $\mathrm{REI}\gtrsim 10$. Velocity-structure analysis of the sample shows that the RPMA $<15^\circ$ subset has stronger positive pairwise expansion than the full co-moving population, providing additional support for the rocket effect. The outward-moving stars are not uniformly distributed; instead, their position angles are strongly anisotropic and preferentially aligned towards the bright-rimmed clouds. Gaia colour-magnitude diagrams show no strong age offset between sources moving outward and literature cluster members. The observed kinematic signatures support feedback-driven acceleration of gas and triggered star formation preferentially towards higher density regions of expanding bubbles due to the rocket effect.
Read the full abstract

All authors R. Arun, G. Maheswar, Annapurni Subramaniam, Piyali Saha, T.P. Sudheesh, B. Shridharan, Blesson Mathew, Akhil Krishna R

Dynamical Systems

Relative ($\tau$), Expanders, and Decay of Correlations for certain Expanding Maps

Relative ($\tau$) is equivalent to a statement that the sequence of Cayley graphs associated to group quotients $\Gamma_q=\Gamma/N_q$, $q\in\mathbb{N}$, form an expander family. There is a philosophy that expander graphs give rise to good mixing; for instance, one has exponential mixing for the geodesic flow uniformly the along a family of congruence covers of the modular surface, stemming from the symmetry in the $\mathrm{SL}(2,\mathbb{R})$ action and the uniform spectral gap for the Laplacian. Do we see similar phenomena in less structured settings? We investigate this question for a tower of finite sheeted covers of certain expanding dynamical systems. We use transfer operator machinery that applies in particular in the cases of subshifts of finite type and expanding interval maps. We make fruitful connections with KMS states of Cuntz–Krieger algebras.
Read the full abstract

Artificial Intelligence

AI for Computational Design Science: A Responsible Human-AI Framework and Case Study on Short-Form Video Safety Surveillance

Artificial intelligence (AI) is transforming not only what information systems researchers design, but also how design research is conducted. Yet existing literature offers limited guidance for computational design science (CDS) when AI actively participates in problem formulation, resource construction, design search, evaluation, and knowledge abstraction. We develop AI for Computational Design Science (AI4CDS), a five-phase methodological framework in which AI expands problem and design search while researchers retain responsibility for domain grounding, admissibility, verification, and scientific judgment. Collaboration is governed by graduated trust, reversibility, auditability, and differentiated reproducibility. We instantiate AI4CDS through ChildRiskGuard, an interpretable artifact for detecting short-form videos inappropriate for children, while documenting AI interactions, rejected alternatives, corrections, and audit trails. The case translates audience-dependent safety and explanation faithfulness into three technical challenges and develops an artifact that separates generic from child-specific risk, represents distinct developmental-risk mechanisms, and makes concept-level explanations part of the predictive computation. ChildRiskGuard achieves an F1 score of 0.769, substantially outperforming direct application of a general-purpose content-safety model while remaining competitive with strong benchmarks. The primary contribution is AI4CDS as a responsible framework for AI-enabled CDS; ChildRiskGuard provides process and artifact evidence of how AI-expanded, researcher-governed design can generate and evaluate novel computational design knowledge.
Read the full abstract

All authors Wenli Zhang, Jiaheng Xie, Zhihe Pan, Yidong Chai, Xiao Fang, Sudha Ram

Cryptography and Security

CONTINUITY: Security-Context Contracts for Composable LLM Agent Controls

LLM agent systems increasingly combine provenance tracking, authorization, policy enforcement, protocol adapters, and execution controls. However, individually correct security mechanisms do not necessarily compose into an end-to-end secure system: security-critical context may be dropped, widened, rebound, or reinterpreted as actions cross component boundaries. We identify this failure mode as security-context discontinuity and introduce CONTINUITY, a framework for verifiable composition of agent security controls. CONTINUITY models each component with an assume-guarantee contract and carries authenticated security context across transitions using signed root grants, provenance commitments, role-bound transition receipts, bounded typed releases, transformation witnesses, and effect-bound execution permits. We formalize end-to-end consequence integrity, requiring every realized external effect to be backed by a valid and current authorization witness linking the principal, task, provenance, delegation, policy state, canonical action, and finality boundary. We implement a reference verifier and deterministic cross-layer fault-injection suite covering 32 fault classes across four application domains. In 2,560 parameterized attack instances spanning 128 fault-domain classes, the full CONTINUITY configuration commits no harmful external effect, while completing all 700 benign tasks and escalating all 200 ambiguous cases. These results show that secure agent execution requires not only sound individual controls, but explicit contracts that preserve their guarantees across the complete instruction-to-effect path.
Read the full abstract

Combinatorics

Ramsey properties of maximal (outer)planar graphs

We study a natural extension of Ramsey theory relative to the classes of maximally planar and maximally outerplanar graphs. This can be seen as a continuation of the study of `Planar Ramsey theory', introduced by Axenovich et al. The question we ask is the following: For a fixed family $\mathcal{K}$ of graphs and a pair of graphs $\{H,F\}$, does there exist an integer $r_{\mathcal{K}} (H, F)$ such that for every graph $G \in \mathcal{K}$ with $|G| \geq r_{\mathcal{K}}(H, F)$, every red/blue edge-colouring of $G$ admits a red copy of $H$ or a blue copy of $F$? When such an integer exists, we say $\{H,F\}$ is unavoidable in $\mathcal{K}$,, and otherwise $\{H,F\}$ is avoidable in $\mathcal{K}$. Our work focuses on this problem where $\mathcal{K} = \mathcal{K}_{\mathrm{MOP}}$ and $\mathcal{K} = \mathcal{K}_{\mathrm{MP}}$, which denote the families of maximal outerplanar (MOP) graphs and maximal planar (MP) graphs, respectively. This framework generalises the classical Ramsey problem relative to these classes, as the case with $\mathcal{K} = \{K_n \colon n \geq 2\}$ corresponds to classical Ramsey. We also study the corresponding Ramsey numbers for MOP and MP, which we denote as $r_{\mathrm{MOP}}(H, F)$ and $r_{\mathrm{MP}}(H, F)$. In the case when $\mathcal{K} = \mathcal{K}_{\mathrm{MOP}}$, we completely determine all unavoidable pairs $\{H, F\}$ with $|E(F)| \geq 2$, together with upper bounds and sometimes exact values of $r_{\mathrm{MOP}}(H, F)$. When $\mathcal{K} = \mathcal{K}_{\mathrm{MP}}$, we completely determine all unavoidable pairs in the diagonal case $\{H, H\}$ when $H$ is connected, showing that $H$ must be one of the graphs $P_3$, $P_4$, $P_5$, $K_{1, 3}$ or the fork graph $S_{2,1,1}$. This work opens up further possibilities in the study of Ramsey theory relative to a class, and we offer several open problems in this vein.
Read the full abstract

Solar and Stellar Astrophysics

Energy Partitioning at the Termination Shock

We show new results of a global 3D magnetohydrodynamic (MHD) simulation of the Boston University outer heliosphere model where we used a newly developed approach that distributes the non-adiabatic shock heating among the cold protons, electrons, and pickup-ions (PUIs), while maintaining total energy conservation of all ions. In previous simulations (E. S. Bair et al. 2025; B. van der Holst et al. 2026) non-adiabatic shock heating was distributed such that the resulting proton temperature jump at the termination shock (TS) was too high. Using a new methodology the simulation results improved with respect to the solar wind proton temperature jump observed at the TS by Voyager 2. Because we directly control the distribution of non-adiabatic heating at the TS, we have the opportunity to study the scenario of a non-adiabatic physical process heating cold plasma and PUIs in the TS and its effect on the global MHD solution. The results show that having a fraction of up to 8.5% of the non-adiabatic shock heating going towards thermal solar wind protons at the TS reproduces the jump conditions observed along the V2 trajectory. This fraction depends on the thermal proton and PUI number densities. This information is key to understanding the physical processes that shape the heliosphere. As shown by M. Opher et al. (2020), PUIs significantly change the shape of the heliosphere; for example, the presence of hot PUIs results in a thinner inner heliosheath. Our work provides the architecture of how energy partitioning at shocks in kinetic simulations (J. Giacalone et al. 2021) can be utilized in global MHD models.
Read the full abstract

All authors Judit Szente, Bart van der Holst, Gabor Toth, Merav Opher

Robotics

TacPAC: Tactile Prediction and Real-Time Action Correction in World-Action Models for Contact-Rich Manipulation

World-action models guide action generation with predicted future observations, but vision-centric predictions miss the local contact cues that decide contact-rich manipulation. However, naively predicting future tactile observations as additional views recovers only a third of the achievable gain in our experiments. This gap reflects a timing mismatch: predictions precede execution, while tactile feedback arrives during it. We introduce TacPAC, which turns tactile prediction into real-time action correction. Once the base model has planned an action chunk, TacPAC caches the predicted contact that plan was conditioned on together with the plan's own representation, and a tactile expert reads each newly observed tactile image against that cache to correct the actions not yet executed. Feedback is thus interpreted against what the plan anticipated rather than in isolation, and one correction is a single pass over that cache, $20.7\times$ cheaper than regenerating the chunk. On five real-robot tasks spanning precision insertion, fragile-object handling, object reorientation, and long-horizon manipulation, TacPAC leads every task and raises the average from 22% for its vision-only base model to 64%. Code is available at https://github.com/LogosRoboticsGroup/TacPAC.
Read the full abstract

All authors Zipei Ma, Xiaofei Wei, Junzhe Jiang, Shunlin Lu, Li Zhang

Earth and Planetary Astrophysics

Enabling Metallicity Measurements of M-dwarf Microlensing Lenses out to the Galactic Bulge

We demonstrate that metallicity becomes identifiable for M-dwarf microlensing lenses out to bulge distances when multi-band lens photometry is combined with the angular Einstein radius, thereby defining the photometric–microlensing metallicity method. The cool atmospheres of M dwarfs contain abundant molecules, whose broad absorption bands make their positions in an optical–NIR color–absolute magnitude diagram sensitive to metallicity. The angular Einstein radius provides the mass–distance constraint needed to infer the lens absolute magnitude, while the intrinsic M-dwarf locus and reddening vector are non-parallel in color–color space, allowing the lens extinction and intrinsic color to be inferred simultaneously from its multi-band photometry, thereby recovering the lens intrinsic position in the metallicity-sensitive color–absolute magnitude diagram. The method requires three-band lens photometry spanning roughly the R, Z, and K bands, making it particularly well suited to Roman through its high-resolution F062, F087, and F213 imaging. With the planned Roman Galactic Bulge Time-Domain Survey (GBTDS) observations plus an additional $\sim$8 hr of F062 imaging per field ($\sim$40 hr in total) roughly a decade after Roman launch, host metallicities could be measured for $\sim$150 planetary systems under the GBTDS yield forecast, with a typical $1\sigma$ precision of $\sim$0.25 dex from our mock-recovery analysis. Applied homogeneously to GBTDS microlensing events with and without detected planets, the method would enable the first measurement of the occurrence–metallicity relation for cold low-mass planets beyond the snow line, constrain the low-metallicity cutoff for their formation, and extend occurrence–metallicity studies into the inner Galaxy.
Read the full abstract

All authors Jiyuan Zhang, Gregory Green, Weicheng Zang, Hongjing Yang, Shude Mao

Earth and Planetary Astrophysics

A Homogeneous Survey of JWST MIRI Transmission Spectra of 10 Exoplanets

The launch of JWST has enabled unprecedented atmospheric characterization of exoplanets. In particular, the JWST-MIRI instrument in the mid-infrared enables probing of molecular species that are harder to detect in the near-infrared. This has come to the fore with the first MIRI-LRS observation of a candidate hycean world, K2-18 b, with potential signs of dimethyl sulfide in its atmosphere. The statistical significance of this result has been debated in subsequent works, arguing instead for the possibility of random noise, instrumental systematics, or a different molecule. The same applies to similar spectra reported for several other exoplanets using MIRI-LRS and interpreted with similar retrieval architectures. In this work, we distinguish between these possibilities with a homogeneous survey of MIRI-LRS transmission spectra of 10 diverse exoplanets, from temperate sub-Neptunes to hot rocky planets and gas giants. We first perform a comparative assessment of the spectra, considering fits with featureless models and pairwise correlations, followed by extensive atmospheric retrievals. With the retrievals, we search for over 150 trace species in the atmosphere of each planet. We find hints of complex molecules in the temperate sub-Neptunes and candidate hycean worlds K2-18 b, TOI-732 c and TOI-270 d, similar to previous works, but not in the other seven planets. Our results indicate that the observed spectral features are likely due to molecular absorption rather than instrument systematics or noise, and underscore these temperate sub-Neptunes as a unique class of planets. More observations are required to robustly identify the specific molecules and constrain the underlying chemical processes, biotic or abiotic.
Read the full abstract

All authors Martin Binet, Nikku Madhusudhan, Måns Holmberg, Subhajit Sarkar

Numerical Analysis

Shallow neural network approximation in mixed Sobolev spaces

We investigate the best $L_2$ approximation of mixed Sobolev spaces by shallow neural networks with $n$ neurons and general activation functions. We first establish an activation-independent Fourier-block principle: if an activation has univariate approximation order $\rho$ in the sense of the Fourier-block property, then the global approximation rate has algebraic order $\min\{\alpha,\rho\}$ for target functions of mixed smoothness $\alpha$, up to explicit logarithmic factors. To verify this property for concrete activations, we introduce a structured univariate approximation condition that implies the Fourier-block property with explicit parameters. For $\mathrm{ReLU}^k$, a matching algebraic lower bound identifies $\min\{\alpha,k+1\}$ as the optimal algebraic approximation exponent in any dimension, up to logarithmic factors in the upper bound. The framework also yields the exponent $\min\{\alpha,k+1\}$ for cardinal B-splines and soft-$\mathrm{ReLU}^k$, and the full mixed-smoothness exponent $\alpha$ for ELU and cosine activations, again up to logarithmic~factors.
Read the full abstract

Computation and Language

Self-Supervised Lexical Representation Learning for Fast, Large-Scale Phylogenetic Inference

Computational phylogenetics has become an essential tool in historical linguistics, yet its application at a global scale remains constrained by two factors: the labor-intensive manual annotation of cognacy judgments required for character-based methods and the substantial computational cost of inference on large datasets. This paper introduces a fully self-supervised contrastive learning framework that learns lexical representations directly from raw IPA-transcribed wordlists, without requiring cognacy annotations, alignments, or additional expert input. The model employs a dual contrastive objective: a word-level loss that organizes phonetically similar forms into a coherent space, and an auxiliary language-level loss that encourages the lexical space to reflect broader phonological properties of languages. From the resulting word representations, pairwise language distances are derived and used to infer a global phylogenetic tree of 3,399 language varieties. The inferred tree achieves a generalized quartet distance (GQD) to the Glottolog reference tree competitive with multiple baselines, while requiring only minutes of computation on a standard notebook GPU. Furthermore, the same representations capture diachronic concept stability: variance in pairwise distances across languages yields stability rankings that correlate significantly with established rankings. Ablation studies confirm that both the language-level objective and the use of phonetic feature vectors improved the inferred trees topology with regards to GQD. The framework thus provides a computationally efficient and fully automatic alternative for large-scale phylogenetic inference and offers a unified representation supporting downstream analyses at both the language and concept level.
Read the full abstract

Artificial Intelligence

Trace2Tower: Transition-Aware EigenTrace Induction of Multi-Level Skills for LLM Agents

Large language model agents increasingly rely on execution traces to master complex interactive tasks. However, current paradigms are bottlenecked by shallow trajectory retrieval and flat skill summarization, fundamentally ignoring the temporal dependencies and outcome-conditioned topology of agent behavior. We introduce Trace2Tower, a transition-aware EigenTrace framework that distills raw trajectories into a robust skill hierarchy. Trace2Tower abstracts step-level interactions into canonical events, constructing a unified graph governed by semantic compatibility, transition dynamics, and outcome evidence. Through a novel contrastive spectral decomposition, it isolates stable, success-aligned behavioral modes while rigorously suppressing failure-prone shortcuts. These modes organically populate a dynamic skill tower of action templates, procedural routines, and overarching task strategies, continuously refined via verifier-guided feedback. On ALFWorld, Trace2Tower achieves 87.31% success requiring only 10.35 steps and 0.26 invalid actions; on WebShop, it reaches 50.67% exact success. Across both benchmarks, Trace2Tower significantly outperforms existing baselines in task mastery and context-efficient experience reuse.
Read the full abstract

All authors Jiazheng Sun, Boyu Yang, Binhao Yuan, Mingxuan Li, Xin Peng

Robotics

One Word, Different Action: A Real-Robot Benchmark for Language-Conditioned Embodied Reasoning

Changes in natural-language instructions can directly alter the behavior ultimately executed by a robot, but such changes do not always imply that the task itself has changed. We propose One Word, Different Action, a language-conditioned executable decision benchmark based on real-robot physical decision anchors. It evaluates robot behavioral responses under task-preserving and task-changing conditions and examines joint reasoning over multiple task constraints. The benchmark uses decision invariance and decision sensitivity to measure action stability when task semantics are preserved and correct decision updates when task requirements change. Matched atomic constituents directly relate correct handling of individual constraints to the corresponding joint executable decision. Experiments show that atomic-task correctness does not sufficiently represent compositional reasoning ability: even when every constituent constraint is handled correctly, a model may still produce an erroneous final action decision on the corresponding compositional task. These results indicate that the key challenge in language-conditioned robot reasoning lies not only in understanding individual task requirements, but also in integrating multiple correctly handled constraints into consistent executable behavior.
Read the full abstract

Combinatorics

Order 14 is the largest order for which every 4-total coloring of every cubic graph is equitable

A total coloring of a graph is an assignment of colors to its vertices and edges so that adjacent or incident elements receive distinct colors, and it is equitable when the cardinalities of any two color classes differ by at most one. Stemock conjectured that every $4$-total coloring of a cubic graph of order less than $20$ is equitable. In this paper, we disprove this conjecture: the circular ladder $L_{12}$ admits a non-equitable $4$-total coloring and, moreover, no smaller counterexample exists: order $4$ is vacuous, and every $4$-total coloring of a cubic graph of order $6$, $8$, or $10$ is equitable. We also prove that the same property holds at order $14$. Our proofs rely on a decomposition lemma, which states that, in any $4$-total coloring of a cubic graph $G$, each color class consists of an independent set $S$ together with a perfect matching of $G-S$. We use the lemma to determine all possible color class configurations for orders $12$, $16$, and $18$, and we show that every listed configuration is attained. Finally, we provide a splicing construction showing that, for every even $n\geq16$, some connected cubic graph of order $n$ admits a non-equitable $4$-total coloring. We may conclude that $14$ is the largest order for which every $4$-total coloring of every cubic graph is equitable.
Read the full abstract

All authors Matheus Adauto, Celina de Figueiredo, Diana Sasaki, Rafael Schneider

Optimization and Control

Ask Before You Optimize: Dynamic Pre-Formulation Clarification for Interactive Optimization

Large language models (LLMs) are increasingly used to formulate optimization models from natural-language problem descriptions, yet realistic operations research (OR) requests are often incomplete: missing objectives, constraints, or business rules can change the resulting mathematical program. Existing evaluations largely assume a complete specification and therefore overlook whether an agent knows when clarification is needed before modeling. We introduce OR-Clarify, a benchmark for pre-formulation clarification. Each task presents a partial public problem description, withholds structured hidden slots, and evaluates agents through bounded interaction with a simulated user. The benchmark supports both openended and choice-based clarification, and measures slot recovery, stopping behavior, silent assumptions, and interaction cost. We further propose Interactive Optimization (InterOPT), a two-stage framework that identifies unresolved formulation-critical gaps and uses them to guide whether to ask the next question or to stop. In our choice-based experiments, InterOPT substantially outperforms all baselines in exact slot recovery; in the open-ended setting, it remains competitive with strong prior methods. Together, OR-Clarify and InterOPT reframe OR assistance as a selective completeness decision: clarify when needed, stop when ready, and quantify what remains missing.
Read the full abstract

All authors Sihan Ge, Yichen Lin, Chenyu Zhou, Jianghao Lin, Tao Yao, Dongdong Ge

Artificial Intelligence

Commonsense Reasoning in Computer Vision: Foundations, Recent Advancements, and Future Directions

Commonsense reasoning in computer vision encompasses integrating visual data and contextual knowledge, crucial for enhancing AI's understanding of everyday scenarios. This understanding not only improves machine learning models but also enhances their ability to interact meaningfully with humans and the environment. Unlike CNN-based conventional vision models, which are designed to identify objects within a specific image, incorporating commonsense knowledge enables models to interpret scenes in a more holistic manner, thereby improving their spatial ability to reason about relationships among objects and actions. This integration not only enhances object recognition but also facilitates a deeper understanding of the contextual factors, ultimately leading to more precise predictions and interactions in real-world applications. This paper presents a comprehensive survey of recent developments that integrate commonsense knowledge into computer vision tasks. We systematically review approaches based on knowledge graphs, scene graphs, neuro-symbolic models, and commonsense-augmented transformers. We also outline current limitations related to dataset bias, knowledge incompleteness, and integration challenges. Finally, we highlight prospective research trajectories in cross-modal reasoning, scalable commonsense knowledge injection, and neuro-symbolic hybrid architectures to develop truly intelligent visual systems.
Read the full abstract

All authors Bahar Uddin Mahmud, Sumit Barua, Guan Yue Hong, Ajay Gupta, Hexu Liu

Signal Processing

Emotion Recognition from Physiological Signals Using Machine Learning Algorithms Under Controlled Emotional Stimuli

Emotion recognition using physiological signals plays a crucial role in well-being analysis, affective computing and human-computer interaction. This study investigates the performance of multiple machine learning models in classifying targets such as discrete emotions with varying granularity, valence and arousal using physiological signals such as Electrocardiogram (ECG) and Galvanic Skin Response (GSR). In here, we extracted various time- and frequency-domain features from the ECG and GSR data to train machine learning models. The results indicate that categorizing discrete emotions with fewer emotions and categorizing arousal achieves good classification accuracy with tree-based models. XGBoost achieved accuracy of 52.8 % for classifying discrete emotions and Random Forest achieved accuracy of 53.6% for classifying arousal. In both the cases, the combination of ECG+GSR feature sets achieved best performance. These findings highlight the effectiveness of physiological signals in capturing emotional states and support their use for emotion recognition systems.
Read the full abstract

Computer Vision

Compact Neural Appearance Models for Efficient Gaussian Splatting

Explicit primitive-based radiance fields such as 3D Gaussian Splatting typically model view-dependent appearance using low-order spherical harmonics (SH). Although efficient to evaluate, SH coefficients dominate per-primitive storage and memory traffic, while their band-limited basis restricts angular detail. We present a thorough, end-to-end comparison of SH and recent spherical appearance models and introduce an implicit alternative that decodes compact per-primitive latent codes using a tiny shared MLP. We integrate all models into the same optimized pipeline, fusing their forward and backward passes into a differentiable CUDA rasterizer and provide a portable WebGL viewer for laptop and mobile GPUs. Our evaluation across reconstruction quality, memory use, and optimization and rendering performance shows that recent spherical models offer the strongest overall quality-efficiency trade-off. Our neural representation is the most compact model evaluated and, compared to third-degree SH, reduces the per-primitive appearance footprint from 192 to 28 bytes, accelerates optimization by 1.3$\times$, while improving reconstruction quality. We further analyze how appearance parametrization shapes optimization, identifying differences in recovered geometry and the tendency of expressive models to absorb non-static scene content. Together, our framework and analysis provide practical guidance for replacing SH beyond what image metrics alone can capture.
Read the full abstract

All authors Florian Hahlbohm, Jorge Condor, Linus Franke, Martin Eisemann, Marcus Magnor

Software Engineering

T(r)opical Islands: Visualizing & Understanding Socio-Technical Artifacts

Projects hosted on collaborative software development platforms, such as GitHub, include many non-code artifacts documenting the project's lifecycle, with its challenges, plans, design, and even community. These socio-technical artifacts include, for example, bug reports, feature requests, and forum posts, offering a useful prospect on the project's evolution. However, these artifacts are dispersed over multiple communication channels and written in natural language, making their analysis difficult, as they are fragmented and with considerable noise. We present a 3D visualization approach mapping topics found across a project's socio-technical artifacts onto vegetation-covered islands, where the individual artifacts are depicted as trees of various types. The topic islands rise out of the ocean as they become discussed, to sink again when they are no longer so. We built a prototype implementing the entire visualization pipeline, from data mining to interactive rendering, leveraging machine learning techniques to cluster the artifacts and extract their topics. We present, through several case studies, the insights that our approach elicits about discussions of development topics throughout a project's history. The user study we conducted (N=34) further strengthens our conclusions about its suitability for understanding socio-technical artifacts and their evolution.
Read the full abstract

All authors Adam Štěpánek, Marco Raglianti, Vít Rusňák, Jan Byška, Barbora Kozlíková, Michele Lanza

Machine Learning

GLASS: Graph-Language Alignment with Spherical Scoring for Transferable Graph-Level Anomaly Detection

We introduce GLASS, a framework for graph-level anomaly detection (GLAD) that achieves robust cross-domain transferability through graph-language alignment on the unit hypersphere. GLASS builds a unified representation space by aligning a structure-aware graph encoder with an instruction-aware text embedding via a multi-slice soft cosine objective. Our framework serializes local, global, and semantic graph properties into a compact Graph Descriptor Prompt (GraphDP), creating a text bridge that enables domain-agnostic anomaly scoring. By enforcing multi-scale consistency through Matryoshka representation slices, the model captures anomalous deviations at multiple levels of granularity. We formulate anomaly detection as density estimation on the aligned hypersphere and introduce Spherical Multi-Modal Scoring (SMS), which instantiates von Mises-Fisher kernel density estimators in both graph and text embedding spaces. This probabilistic formulation recovers angular 1-nearest-neighbor scoring in the high-concentration limit, motivates the practical mean k-nearest-neighbor scorer, and provides a principled fusion of structural and semantic anomaly signals. The shared text embedding space further serves as a cross-domain bridge: by encoding a target domain's GraphDP without target-domain training data, GLASS performs zero-shot anomaly detection, and with only a handful of normal examples, few-shot adaptation via reference-set calibration. For privacy-sensitive deployment, we extend reference-set calibration with a bounded joint graph-text kernel summary that provides graph-record differential privacy while keeping the encoders fixed independently of the private target references. Across twelve benchmarks and three meta-domains, GLASS obtains the best average AUROC and rank compared with recent advanced GLAD baselines and enables effective cross-domain transfer.
Read the full abstract

All authors Xudong Wang, Chris Ding, Tongxin Li, Jicong Fan

Algebraic Geometry

Coleman Isomorphisms in Syntomic Cohomology and $\mathrm{THH}$

This paper proves a generalisation of Coleman's isomorphism (between norm compatible cyclotomic units and a group of invertible power series) to various cohomology theories evaluated on proper regular $p$-adic formal schemes, for future applications to Iwasawa theory. This generalisation is an immediate consequence of a description, as a cyclotomic synthetic spectrum, of the limit over transfer maps as one goes up the cyclotomic tower of motivically filtered $\mathrm{THH}$. This crucially uses a calculation of $\mathrm{THH}(\mathbb{Z}_p[\zeta_{p^n}])$ due to Devalapurkar–Raksit as well as a calculation of the free loop transfer due to Schlichtkrull. Along the way, we construct transfer maps for various cohomology theories along finite locally free regular maps, using some elements of $\mathbb{P}^1$-stable motivic homotopy theory.
Read the full abstract

Artificial Intelligence

A Unified Physics-Aware Quantum Machine Learning Framework across Power GaN HEMTs and Logic Nanowire FETs: Predicting Unseen Process Splits and Held-Out Geometry Combinations with Lower Error and Tighter Split-to-Split Variability

We present a unified reinforcement-learning (RL) framework that discovers compact parametrized quantum circuits (PQCs) for data-scarce device modeling. A graph neural network (GNN) policy optimized by proximal policy optimization (PPO) searches circuit architectures using leave-one-group-out cross-validation (LOGOCV) error on held-out process or geometry groups as the reward. The framework achieves the lowest mean absolute error (MAE) on all 11 targets versus six classical baselines, with 59% lower error (Ioff) and 81% tighter fold variability (VTH) for HEMTs and 84% lower error (VTH, SS, Ioff) and 82% tighter fold variability (Ioff) for NWFETs. These results demonstrate the potential of RL-selected, classically simulated PQCs as compact surrogates with low OOD error and improved physical consistency, despite imposing no explicit physical constraints, penalty terms, or device-specific equations, on the two evaluated device datasets.
Read the full abstract

All authors Rushat Rai, Yun-Yuan Wang, Autsada Kakaen, Pei-Jie Chang, Doan Viet Nguyen, Yuan-Chieh Chiu, Doldet Tantraviwat, Niall Tumilty, Simon See, Wen-Jay Lee, Tai-Yue Li, Nan-Yow Chen, Tian-Li Wu

Quantum Physics

Memory-Optimal Sequential Synthesis of Multimode Gaussian Transformations

In modular quantum computing architectures, communication between hardware modules is mediated by traveling qumodes sent through transmission lines. Each output qumode interacts with the emitting module only once through a beam-splitter-type interaction and becomes inaccessible to that module after emission. Information required for subsequent outputs must therefore remain in long-lived memory qumodes. For a prescribed multimode Gaussian transformation on $N$ qumodes, this work determines the minimum memory cost for any given emission order, constructs an explicit sequential protocol attaining this minimum, and develops a greedy method for identifying memory-efficient emission orders. The transformation is represented by a symplectic matrix $S$, specified either directly or through a Gaussian gate sequence. The exact minimum memory cost is obtained from the ranks of submatrices of $S$ and further reduces to a support-based counting rule whose computational cost is linear in the size of the support data. When $S$ is specified directly, a matrix-based protocol attains the minimum memory cost. If instead $S$ is specified through a gate sequence, the original gates can be reused without additional synthesis, although the resulting memory usage need not be minimal. Gaussian transformations with local support on a $D$-dimensional cubic lattice can be realized sequentially with $O(N^{(D-1)/D})$ memory qumodes. The protocols also apply to non-Gaussian inputs, including GKP and cat states, and thereby provide an explicit, resource-efficient scheme for intermodule communication in modular architectures for universal continuous-variable quantum computation.
Read the full abstract

Hardware Architecture

Proton Irradiation Characterization of an Open-Source ML Accelerator on a Zynq UltraScale+ MPSoC

As spaceborne computing systems increasingly rely on neural network (NN) accelerators, the opacity of commercial, black-box architectures severely restricts the development of verifiable radiation mitigation strategies. Open-source, register-transfer level (RTL)-accessible accelerators resolve this limitation by enabling user-defined instrumentation, yet few have empirical radiation-response baselines. This work establishes a foundational system-level proton-irradiation baseline for an unmitigated open-source Tensil NN accelerator deployed on a Zynq UltraScale+ SoC executing ResNet-20 inference. Under 20 to 58 MeV proton irradiation, we delivered $4.29 \times 10^{10}$ p/cm$^{2}$ within monitored operational windows. Seven workload interruptions required two restarts of the notebook process, four reboots or board resets, and one power-cycle sequence. Two output-corruption events returned incorrect CIFAR-10 classes without loss of service. In the longer event, the accelerator returned a class absent from the ten-image CIFAR-10 pool for 39 consecutive inputs at normal cadence. The process remained alive, while the kernel log, limited memory test, and sampled power showed no anomaly. Observation of the stuck-class sequence ended with scheduled bitstream reconfiguration. All nine onsets occurred under the nominal 4 cm beam, which exposed the SoC, LPDDR4, and additional board circuitry; none occurred under the 2 cm SoC-centered field. This pattern shows a field association but does not establish LPDDR4 as the cause because field size was confounded with run order and dose. Linux-managed accelerators require end-to-end content checks and recovery that reaches the state in which corruption can persist. This baseline documents availability loss and silent output corruption, supporting future software hardening of COTS FPGA-SoCs for neural-network inference in space systems.
Read the full abstract

All authors Saad Memon, Rafal Graczyk, Jan Swakoń, Leszek Grzanka, Sebastian Kusyk, Mike Papadakis

Quantum Physics

Quantum-State Texture Dynamics: Theory and Experiment

Quantum-state texture (QST) has found applications in several fields from quantum foundations and computation to quantum criticality. However, a general theory of QST dynamics under arbitrary physical processes remains unavailable, limiting both its practical application and experimental exploration. Here, we demonstrate that the QST response to an arbitrary finite-dimensional channel is fully encoded in the dual evolution of a single reference state. This description yields necessary and sufficient conditions for texture preservation and implies exact conservation under all free-unital dynamics. Using a nuclear magnetic resonance quantum processor, we experimentally verify these predictions across distinct channel classes. Furthermore, we show that local QST measurements provide an operational signature of entangling gates in circuit layers. Our results establish quantum-state texture as a resource and a practical diagnostic tool in quantum information processing.
Read the full abstract

All authors Carlos H. S. Vieira, Xinfang Nie, Dawei Lu, Fernando Parisio

Statistical Mechanics

Diffusion under competing bulk and surface stopping mechanisms

We investigate reflected diffusion in a bounded domain subject to two independent, competing stopping mechanisms: an exponentially distributed bulk lifetime of rate $p$ and a surface reaction triggered when the boundary local time exceeds an independent exponential threshold of rate $q$. Denoting by $T$ the stopping time and by $L$ the acquired boundary local time at stopping, we derive their marginal distributions, joint Laplace transform, and complete hierarchy of mixed moments. These statistics are determined by the splitting probability $\phi$ that surface reaction occurs before bulk decay. In particular, we establish the identity $p\expect{T}+q\expect{L}=1$ and show that the cumulative risk $pT+qL$ is exponentially distributed with unit rate. We further obtain equivalent representations of $\phi$ in terms of the Robin-Laplacian and the generalized Steklov spectra. Explicit results for a three-dimensional ball reveal how competing rates $p,q$ control $\phi$ and the $(T,L)$ statistics. Monte Carlo simulations test the universal cumulative-risk law.
Read the full abstract

Numerical Analysis

A Generalized Scalar Auxiliary Variable Method for Structure-Preserving and Efficient Integration of Nonlinear port-Hamiltonian DAEs

We develop an energy-optimal generalized scalar auxiliary variable (EOP-GSAV) framework for nonlinear index-one port-Hamiltonian differential-algebraic equations (pH-DAEs). Exploiting the port-Hamiltonian structure, we separate the nonlinear effort, interconnection, and dissipation terms from a constant implicit core. The resulting BDF-1 and BDF-2 schemes require one linear solve per time step with a reusable factorization, while retaining discrete passivity and accurate tracking of the Hamiltonian. The schemes are compared with the implicit midpoint method equipped with full, modified, and frozen-Jacobian Newton iterations. Numerical experiments ranging from a strongly state-dependent nonlinear stress test to large-scale benchmarks demonstrate robust and competitive performance, with substantial efficiency gains in matched-accuracy regimes. A comparison with SUNDIALS IDA shows comparable work-precision behavior at equal order despite a non-specialized Python/SciPy implementation, while unrestricted variable-order adaptive IDA is faster in the high-accuracy regime.
Read the full abstract

Artificial Intelligence

Do LLMs Exhibit Coherent Knowledge Structures in Mathematical Reasoning? A Perspective from Knowledge Space Theory

Human knowledge is inherently structured and interdependent: mastery of a concept requires prior mastery of its prerequisites, a principle formalized by Knowledge Space Theory (KST). While LLMs achieve strong performance on complex reasoning tasks, it remains unclear whether they exhibit coherent, human-like knowledge structure. We introduce a KST-grounded framework for evaluating LLM knowledge structure in mathematical reasoning, using it as a normative framework to analyze whether LLM behavior adheres to principled knowledge dependencies. Evaluating eight open- and closed-source LLMs against real human learners, we find that (1) LLMs do not adhere to human knowledge structure – they frequently violate knowledge dependencies and fail to leverage related knowledge provided in context to improve performance on dependent questions; (2) LLMs do not share a consistent knowledge structure among themselves, as reflected by low overlap in their knowledge distributions. Furthermore, these structural deficiencies remain largely invisible to accuracy-based and LLM-as-judge evaluations. Together, our results provide behavioral evidence that current LLMs knowledge does not follow a human-like structure.
Read the full abstract