Peer Revenue Returns subject to peer review

Papers

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

All of arXiv

Monday 31 August

Artificial Intelligence

Validity-Aware Jailbreak Evaluation for Large Language Models

Jailbreak robustness has become central to large language model (LLM) safety evaluation, yet prevailing methodologies rely primarily on refusal behavior, semantic resemblance, and intent-matching heuristics that emphasize linguistic plausibility rather than correctness. We identify a key limitation in existing evaluations: many jailbreak intents depend on instructional validity rather than epistemic factuality, allowing realistic-looking responses to be labeled successful despite being factually or procedurally incorrect. To address this gap, we propose Sequential Epistemic and Action-Level Validation (SEAV), a verification-centric jailbreak evaluation framework that decomposes responses into ordered steps and evaluates both validity and correctness. SEAV combines LLM-as-a-judge mechanisms for semantic interpretation with retrieval-grounded verification using external knowledge sources, assessing whether generated content is factually correct, structurally consistent, and operationally capable of advancing harmful objectives. Empirically, SEAV cuts the false-positive rate on SD-A (a curated strategic-dishonesty diagnostic) by 14.9\,pp vs. the strongest baseline, and reclassifies 22.1%–51.0% of sampled prior-labeled successes as invalid across three of four public benchmarks. Together, these results show that enforcing correctness substantially reshapes measured robustness: many previously labeled jailbreak successes are reclassified as invalid, and results are stable across the tested search backends and evaluator models. Code and data are available at https://github.com/Ardor-Wu/SEAV.
Read the full abstract

All authors Qilong Wu, Sahil Wadhwa, Pranab Mohanty, Giri Iyengar, Varun Chandrasekaran

Systems and Control

Decentralized and Equilibrium-Set-Oriented Stability Analysis and Control for Power Systems

Conventional power-system stability analysis is largely centralized and centered on a single equilibrium point, which becomes increasingly restrictive in the presence of large-scale fluctuating renewable generation. This paper develops a decentralized framework for stability analysis and control that certifies the asymptotic stability of an equilibrium set rather than that of a given single operating point. To this end, we introduce a new notion termed input–output differential passivity (IODP), which decomposes equilibrium-set stability of the interconnected system into local requirements imposed on individual devices. These requirements are formulated without embedding a particular operating equilibrium into the local conditions; once the certified regions are constructed, stability verification for a given operating scenario reduces to checking whether its equilibrium lies in the certified set. The proposed conditions require each bus to possess a sufficient level of IODP, quantified by an IODP index. To compensate for an IODP shortage, we further develop an I/O-transformation-based passivation controller that reshapes the local input–output behavior of the corresponding device. In this way, all grid-connected components can be made to satisfy the decentralized conditions for system-wide stability. The proposed framework is validated on a modified IEEE 39-bus system. Simulation results demonstrate that it provides a scalable and equilibrium-set-oriented solution for stability certification and control under highly variable operating conditions.
Read the full abstract

Astrophysics of Galaxies

Detection of a triple nucleus in the Wolf-Rayet composite Sy 2 galaxy NGC 6764

We report adaptive-optics Ks-band imaging of the composite Seyfert 2 and Wolf-Rayet galaxy NGC 6764 (D = 32 Mpc), obtained with LUCI/SOUL at the Large Binocular Telescope. The nucleus resolves into three compact sources with pairwise projected separations of 0.07-0.20 arcsec, corresponding to 11.0-30.9 pc. All three have indistinguishable H, Ks, H2, and Br-gamma colours, so photometry alone cannot separate accretion from star formation. The system is consistent with a Seyfert nucleus and two star-forming regions, but a dual or triple AGN cannot be excluded. In the latter case, the separations would be two orders of magnitude smaller than in any confirmed or candidate AGN triplet reported to date. Diffraction-limited near-infrared integral-field spectroscopy is required to establish the nature of each component.
Read the full abstract

All authors Jochen Heidt, Francisco Pozo Nuñez, Nina Mackensen, Magdalena Vukojevic, David Thompson

Computation and Language

Beyond Token Positions: Safety Alignment Across Denoising Steps in Diffusion Language Models

Diffusion large language models (dLLMs) generate text through iterative denoising rather than left-to-right decoding. This generation paradigm introduces two axes that can influence safety alignment: when tokens are generated during denoising and where they appear in the response. In this paper, we measure dLLM safety behavior under harmful prompts by tracing intermediate token distributions and commitment decisions throughout denoising. Our analysis shows that refusal signals are concentrated in early denoising steps and leading response positions, and the tokens committed early can strongly shape the final safety outcome. Our measurements further show that the denoising step and persistence of refusal-token commitment are important for understanding dLLM safety. Based on these findings, we propose Refusal-Aware Early Commitment (RAEC), a simple training-free decoding method that commits persistent refusal signals from early steps. Experiments on LLaDA and Dream show that RAEC reduces attack success rates while largely preserving utility. The code is available at https://github.com/Glresearch1/RAEC.
Read the full abstract

All authors Guoli Wang, Haonan Shi, Tu Ouyang, An Wang

Computation and Language

Human-Anchored Factuality Evaluation with Strategic Annotation

LLM-based factuality judges provide scalable evaluation signals, but their metrics are often systematically biased relative to human judgments. We study human-anchored factuality evaluation under limited annotation budgets, where judge predictions on the full dataset are combined with human labels on a small selectively sampled subset to obtain statistically valid estimates. The efficiency of this approach depends critically on which examples receive human annotation: in factuality evaluation, judge-human misalignment is not driven solely by low confidence, but also by structured failure modes such as incomplete evidence, temporal mismatch, unverifiable claims, and rubric misalignment. To exploit this structure, we introduce a factuality-specific annotation policy design pipeline that uses failure-space analysis (FSA) to derive diverse predictive signals for modeling human-judge misalignment. On an internal reference-based factuality evaluation system (AutoFA) and RAGTruth, where judge-predicted estimates substantially underestimate human-annotated factual accuracy, our FSA-guided policy improves annotation efficiency over uniform sampling and uncertainty-driven baselines, achieving effective-sample-size gains of 40.3% on AutoFA and 27.1% on RAGTruth.
Read the full abstract

K-Theory and Homology

Higher Kazhdan projections for one relator groups

We study the $K$-theory classes of higher Kazhdan projections of a discrete group under a spectral-gap assumption and recall how they are obtained, via the Baum–Connes assembly map, from equivariant Euler classes. After reviewing the case of virtually free groups, where the universal space for proper actions has a one-dimensional model, we prove the main new result: an explicit description of the $K$-theory class of the first higher Kazhdan projection for finitely generated infinite one-relator groups whose first Laplacian has a spectral gap. If moreover the one-relator group is hyperbolic, this also gives corresponding formulas for delocalised $\ell^2$-Betti numbers.
Read the full abstract

Artificial Intelligence

The Privacy-Hallucination Tradeoff in Differentially Private Language Models

Both privacy and factual accuracy are paramount in high-stakes domains like healthcare. Concerningly, we uncover and investigate a privacy-hallucination tradeoff in differentially private (DP) language models. First, we empirically show that models pre-trained or fine-tuned with DP tend to produce more hallucinations than non-DP counterparts, with increased severity as the privacy budget grows stricter. Second, we investigate model properties driving this tradeoff, demonstrating that DP mechanisms flatten output distributions, potentially redistributing probability mass toward factually incorrect alternatives. Third, through experiments where we control fact frequency in training data, we characterize how information frequency can reduce hallucination risks in DP models. Overall, our findings underscore the need for more nuanced privacy-preserving interventions that offer rigorous privacy guarantees without compromising factual accuracy.
Read the full abstract

All authors Krithika Ramesh, Krishna Pillutla, Danish Pruthi, Anjalie Field

Computation and Language

MemeBridge: A Dataset for Benchmarking and Mitigating the Bidirectional Cultural Gap in Meme Interpretation

Communicating across cultures is inherently challenging, especially through culturally dense and ambiguous formats like memes. While people expect large language models (LLMs) to hold promise for bridging such gaps, existing benchmark datasets often fail to capture the cultural context necessary for accurate interpretation. To address this, we introduce MemeBridge, a curated dataset centered on U.S.-originated memes, designed to capture two complementary perspectives: (1) how Chinese participants interpret these memes, and (2) how U.S. participants anticipate how people from other cultures might misunderstand them. Here, context refers to implicit cultural knowledge, including background beliefs, norms, and shared assumptions that shape meme comprehension. The dataset was constructed via a multi-stage crowdsourcing pipeline with rigorous validation, including human agreement checks and GPT-based classification verification. Each meme is annotated with sentiment, emotion, cultural significance, and knowledge type, providing rich supervision for downstream tasks. Notably, we observe that the anticipated misunderstandings from U.S. participants are often inaccurate, highlighting the asymmetries in cultural understanding and the challenges of adopting perspectives beyond one's own. This bidirectional framing, which focuses on both expression and perception, enables more nuanced benchmarking of cross-cultural comprehension. Our probing of multiple LLMs reveals that while models developed in different cultural contexts exhibit partial cross-cultural understanding, they often struggle with sophisticated interpretations. By contrast, fine-tuning with MemeBridge improves model performance, underscoring the value of culturally grounded resources for training and evaluating LLMs in globally diverse settings.
Read the full abstract

All authors Hangxiao Zhu, Suliu Qin, Zhuoyan Li, Ming Jiang, Yu Zhang, Meng Xia

Dynamical Systems

Entropy stability for diffeomorphisms isotopic to Anosov on $\mathbb{T}^d$

In this paper we study the behavior of topological entropy for partially hyperbolic diffeomorphisms on $\mathbb{T}^d$ isotopic to a linear Anosov automorphism with indecomposable weak-stable subspace. In particular, we prove that for a class of DA maps associated with such linear systems, whose central behavior is sufficiently dominated by the expansion rate of the linear model, the topological entropy coincides with that of the linear map. Furthermore, we obtain uniqueness of equilibrium states for a class of low-oscillation potentials and, as an application, the uniqueness of the measure of maximal entropy. We also provide sufficient conditions for which ergodic measures arise as the unique equilibrium state of some continuous potential. Finally, on $\mathbb{T}^3$ we construct a dynamically coherent diffeomorphism in the same isotopy class whose topological entropy is strictly larger than that of the linear part.
Read the full abstract

Machine Learning

A hybrid quantum-classical neural network for learning to route

This work studies hybrid quantum-classical neural networks for learning routing heuristics. Specifically, this paper asks whether small quantum neural networks can replace parameter-heavy modules inside a competitive attention-based routing model while maintaining solution quality. For the capacitated vehicle routing problem, encoder feed-forward replacement emerges as the most promising design: it reduces the number of model parameters by 56.6% while keeping the hybrid model close to the classical neural baseline at small and medium instance sizes, although the gap grows for larger instances. This work also compares to classical routing algorithms, which remain highly competitive and often superior on the fixed Euclidean test sets. Our results therefore do not indicate quantum advantage or solver dominance, but identify encoder feed-forward replacement as a viable hybrid-module compression strategy for neural combinatorial optimization.
Read the full abstract

All authors Marcus Rolf Peter Ritt, Alexsandro Santos da Rosa Júnior, Marcos Vinicius Reballo, Cesar Augusto do Amaral, Fernando Augusto Caletti de Barros

Machine Learning

AdaptNTK: Adaptive Uncertainty Quantification and Active Learning for Neural Network Potentials

Machine learning interatomic potentials bridge the gap between quantum chemical precision and classical computational speed, enabling molecular dynamics simulations with first-principles accuracy. Their reliability is often improved through active learning, which iteratively expands the training set by identifying uncertain, out-of-distribution configurations. Existing uncertainty-quantification methods often involve a trade-off between computational cost and reliability, and generally cannot account for redundancy as an acquisition batch is assembled. Here, we introduce AdaptNTK, a single-model framework that measures uncertainty as a regularized Mahalanobis distance in empirical neural tangent kernel (NTK) feature space. With the NTK features fixed during acquisition, the uncertainty depends on the acquired configurations but not their reference labels. This allows the uncertainty to be updated recursively after each selection without retraining, reducing redundancy within an acquisition batch. On held-out rMD17 data, AdaptNTK achieves the highest mean correlations with force errors (Spearman 0.68, Pearson 0.71) and matches a three-member ensemble in error retention. In active learning experiments, AdaptNTK achieves the lowest force errors across rMD17 and Transition-1X, with particularly strong performance on transition-state configurations in Transition-1X. AdaptNTK provides a 2.6-fold speedup per Transition-1X cycle relative to the ensemble, providing efficient single-model uncertainty estimation with sequential updates for data-efficient active learning.
Read the full abstract

Computation and Language

EvoFlint: An Evolutionary Atlas of Multi-Turn LLM Vulnerabilities

Frontier language models that refuse harmful single-turn prompts often comply when the same intent is reached gradually over many turns, making multi-turn attacks one of the least understood failure modes of large language models. Most automated red-teaming methods treat this as a generation problem: produce attacks that break the model. We argue it is better framed as a search problem: discover, organize, and iteratively refine a diverse archive of attack strategies, producing a structured map of how a target model fails rather than a list of one-off successes. We introduce EvoFlint, which applies evolutionary quality-diversity search to multi-turn red-teaming. Attack strategies are phased conversation plans, not raw prompts, and are evolved through LLM-driven mutation and crossover. A Pareto fitness over attack success rate and peak severity preserves selection signal from near-miss attacks. A risk-indexed archive runs novelty search with local competition over strategy description embeddings inside each cell, maintaining diversity without committing to a predefined style taxonomy. A generation-level memory accumulates target-model insights across the population and feeds them back into strategy generation. On the HarmBench-test split, EvoFlint reaches attack success rates of 35.8% on Claude Sonnet 4.6, 59.7% on GPT-5.4, and 94.3% on Qwen3-32B, alongside 98.7% on the older GPT-4o included as a baseline reference. The resulting archive, organized by risk category, exposes for each target which categories of harm its safety training has and has not covered.
Read the full abstract

All authors Feitong Qiao, Liren Peng, Shiming Ren, Aishwarya Jadhav, Arghavan Bahadorinejad, Marinette Chen, Muhan Zhang, Abdulaziz Suria, Gennevi Lu, Anish Das Sarma

Game Theory

Nash Core in Multiwinner Election

In the approval-based committee selection problem, a committee is said to be in the core if no subset of voters has an incentive to deviate by selecting a blocking committee of proportional size, such that every voter in the deviating group strictly prefers the blocking committee. We consider the setting where candidates can be selected fractionally. Under a mild regularity assumption, we show that there always exists a weighting of candidates such that the fractional committee maximizing the candidate-weighted Nash Social Welfare is in the core. We refer to such a solution as being in the Nash core. Additionally, we show that a Nash core solution admits a payment assignment between voters and candidates, where each voter pays a candidate they approve in proportion to the weight. For the discrete setting, where each candidate is either included or excluded from the committee, we prove that every approval-based committee election with at most eight equally weighted voters has a core committee by rounding the fractional Nash core solution. Although the non-emptiness of the core in this setting remains an open question and checking core membership is coNP-hard, we extend the notion of the Nash core to the discrete case, yielding a formulation that is efficiently verifiable and offers a promising path toward establishing core existence in discrete settings. Finally, we test our approach on real voting data using a payment-guided heuristic. We empirically show that the Nash core solution can be efficiently computed through an iterative algorithm in both the fractional and discrete settings.
Read the full abstract

General Relativity and Quantum Cosmology

Dynamical evolution of quantum mutual information in Schwarzschild spacetime

Quantum mutual information is a fundamental quantity for characterizing correlations among quantum systems. In this work, we investigate the dynamic evolution of the quantum mutual information based on Rényi-2 entropy for three-mode Gaussian states in the background of a Schwarzschild black hole. We find that the physically inaccessible mutual information increases with Hawking temperature and eventually saturates, while the physically accessible mutual information exhibits nonmonotonic temperature dependence: it first rises, reaches a peak, then declines toward a finite asymptotic value. This nonmonotonic behavior differs from the monotonic degradation typically observed for Gaussian entanglement and steering in curved spacetime. Furthermore, we establish several constraint relations governing the distribution of mutual information among subsystems. These findings contribute to understanding the quantumness of quantum correlations for continuous variable in Schwarzschild spacetime.
Read the full abstract

All authors Guosen Ma, Fang Rao, Jingjing Hua, Xiaofen Huang

Optimization and Control

Extremum Seeking Control: Three Revolutions and the Road Ahead

The history of extremum seeking is not merely the history of an algorithm; it is the history of an idea. Few ideas in control engineering have demonstrated the remarkable longevity of extremum seeking control (ESC). Invented more than one hundred years ago, ESC has continually reinvented itself while remaining faithful to its original objective: enabling systems to optimize their performance without relying on accurate mathematical models. This perspective article proposes that the evolution of ESC is best understood through three scientific revolutions. The first Engineering Revolution (1922-1999) established the engineering principles of model-free optimization; the second Mathematical Revolution (2000-2010) provided the rigorous mathematical foundations that transformed ESC into a mature discipline of nonlinear control; and the third ongoing Infinite-Dimensional and Cyber-Physical Revolution (2010-present) continues to expand its scope toward delays, partial differential equations, distributed optimization, event-triggered implementations, and increasingly complex cyber-physical systems. Beyond recounting this historical evolution, we offer a personal perspective on why ESC has remained relevant across successive technological eras. We argue that its enduring influence arises because the fundamental engineering challenge has never changed: "how can a dynamical system learn to improve its own performance when the optimum is unknown?" As optimization, learning, and feedback control become increasingly intertwined, ESC appears uniquely positioned to contribute to a new generation of intelligent autonomous systems, pointing toward what may become the field's fourth scientific revolution.
Read the full abstract

Differential Geometry

On Embedding Hamming Spheres with Applications to Positive Curvature

We consider isometric $\mathbb{Z}_{p}$-torus actions on closed, positively curved manifolds, obtaining a homotopy classification result. This allows us to strengthen on the improvement of the $3n/8$ half-maximal symmetry rank result of Fang-Rong and Ghazawneh given in Theorem B and Corollary C in the work of the author and Searle. Along the way, we generalize a sphere-embedding construction of Wilking from the binary case to arbitrary primes in Theorem C, which is a geometric construction of independent interest.
Read the full abstract

Computation and Language

Are Near-Tied LLM Rankings Robust to Family-DIF-Guided Benchmark Recomposition?

Small leaderboard gaps are often interpreted as evidence that one language model is better than another, but their sign may depend on which benchmark items are included. We test this using item-level responses from five benchmarks and a family-label-free spectral approximation to multidimensional item-response theory (MIRT). In owner-disjoint folds, one owner half identifies items with low residual differential item functioning across model families (low-DIF). These items are used to score models in the other half with frozen weights that preserve the benchmark's composition across metadata-defined item groups and easiness strata. Equally short matched-random subtests provide a baseline for variation due to item subsampling. Full-benchmark and low-DIF rankings remain strongly correlated ($\tau_b=.900$–$.948$). Yet in four of five benchmarks, 30.9–47.1% of cross-family pairs initially within one percentage point reverse order, exceeding their matched-random medians by 16.9–28.6 percentage points (all $p=.001$). The fifth benchmark shows no reliable excess ($-0.9$ points, $p=.689$). The pattern survives all pre-specified population perturbations, and residual item–family signatures replicate across owner halves; however, no family shows a consistent advantage across benchmarks. Thus, globally stable rankings can still leave individual near-tie orderings sensitive to benchmark composition, and sub-one-point leaderboard gaps should be accompanied by evidence that the implied ordering is composition-robust.
Read the full abstract

Classical Analysis and ODEs

The frameset of the second-order B-Spline along $ab=1/q$

We give a complete characterization of the zero set of the Zak transform of the second-order B-spline $Q_2$. As a consequence, we obtain a new proof of the Lemvig–Nielsen obstructions to the Gabor frame property of $Q_2$, and, along the hyperbola $ab=1/q$, we show these obstructions are sharp: every point they do not exclude belongs to the frame set. This gives the first complete characterization of $\mathcal F(Q_2)$ along an entire curve in this family, resolving a question left open by Lemvig and Nielsen.
Read the full abstract

Human-Computer Interaction

Design principles to Increase Technology Self-efficacy for Older Australians with Mild Cognitive Impairment (MCI) and Older Carers

The number of people with age related physical or cognitive impairments is increasing due to the worlds ageing population. Technology has the potential to support independent living, and to achieve aged and health service efficiencies, however, there are gaps in our understanding of factors that motivate technology adoption and ongoing use by older adults, especially those with cognitive impairments. This study aims to explore motivators and enablers for technology adoption and ongoing use by older Australians with mild cognitive impairment (MCI) and their carers, to identify technology design principles and guidelines that maximise adoption. Semi structured interviews were used to gather data about individual demographics, needs, priorities, lifestyle, challenges, and experiences with technology. Results of inductive, reflective, thematic analysis indicate that a desire for independence, autonomy and quality of life motivate use of technology, and perceived technology self-efficacy and IT literacy are enablers. The Protection Motivation Theory illustrates that constant technology change is a disabler for technology adoption and sustained use, because it lowers perceived technology self-efficacy and IT literacy, and reduces confidence to use technology. Two high-level technology design principles and related guidelines are proposed, grounded in theory and aligned with Banduras four sources of self-efficacy. These design principles and guidelines are intended to increase feelings of self-efficacy and confident use of technology, while also reducing adverse impacts of technological change and encouraging sustained technology adoption by older adults with MCI to support independent living and quality of life.
Read the full abstract

All authors Snezna Bizilj Schmidt, Nathan D'Cunha, Stephen Isbel, Blooma John, Ram Subramanian

Artificial Intelligence

EGT-KG: Evidence-Grounded Typed KG Retrieval for Practical Scientific QA with Small Language Models

For emerging scientific research domains, local Small Language Models (SLMs) are becoming more attractive, as they offer stronger privacy control and more stable deployment pipelines than Large Language Models. However, in practice, scientific question-answering on SLMs often operates under inevitable constraints: small literature collections, fragmented evidence, limited context window and reasoning abilities. We propose the Evidence-Grounded Typed Knowledge Graph (EGT-KG), a retrieval framework to improve information retrieval with local SLMs. We assessed three question-answering settings: a vanilla Retrieval-Augmented Generation (RAG) workflow and two EGT-KG workflows: an automatically generated relation schema (AS) and an expert-defined relation schema (ES). Our experiments were evaluated with a six-dimensional evaluation framework (S3CRF: Soundness, Correctness, Completeness, Conciseness, Relevance, Fluency) on a Biopolymer-bound Soil Composite literature benchmark, showing that EGT-KG outperforms the vanilla RAG method in most settings, with the best improvement from llama3:8b: a Final Score of 70.37 (+14.67%) and 68.82 (+12.14%) by AS/ES EGT-KG variants.
Read the full abstract

All authors Muran Yu, Jiechao Gao, Yuandong Pan, Barney H. Miao, Andrew C. Lesh, Kincho H. Law, Jie Wang, Michael D. Lepech

Robotics

Beyond Object Selection:Markerless Gaze-based Robot Placement at Arbitrary Positions

Gaze-based assistive manipulation typically supports object selection, while arbitrary-position placement requires accurate spatial alignment between the headset and robot. However, for gaze-based manipulation, pose accuracy does not necessarily translate into task accuracy: translational and rotational errors jointly affect the transformed gaze ray and may compensate for each other. To study cross-device alignment from this task-oriented perspective, we present a markerless interaction framework and a dedicated cross-device dataset. We propose Graph-based Reference Selection to address sparse robot references. We further develop and benchmark multiple task-specific alignment pipelines under a unified protocol. Specifically, we introduce Gaze–Surface Intersection Error (GSIE), which directly measures the spatial error of the gaze-specified target. Experiments show that alignment methods ranked highly by conventional pose metrics are not always optimal in GSIE, demonstrating the importance of evaluating gaze-based manipulation at the task level.
Read the full abstract

All authors Yuzhi Lai, William Marx, Shenghai Yuan, Peizheng Li, Zhuoyu Ran, Andreas Zell

Solar and Stellar Astrophysics

SPIRITS 19q: Dust Production by a Subsolar-metallicity Carbon-rich Wolf-Rayet Star in NGC 2403

We present JWST/NIRSpec IFU observations of SPIRITS 19q, the highly dust-producing carbon-rich (WC) binary candidate located in a subsolar-metallicity region of the nearby spiral galaxy NGC 2403. The observations, taken in April of 2024, confirm the association of a dusty outburst observed in 2019 by the Spitzer Space Telescope with an early-type WC star. Using models from the Potsdam Wolf-Rayet (PoWR) LMC model grid we find that the WC star of SPIRITS 19q likely has an especially high mass-loss rate ($\gtrsim$ 10$^{-4}$ $M_{\odot}$ yr$^{-1}$). From the flux peak of the IR transient as measured by Spitzer/IRAC as well as constraints on dust composition and size from the JWST spectrum, we estimate a total dust mass formed in the outburst of 6.6 $\pm$ 0.4 $\times$ 10$^{-6}$ $M_{\odot}$. Assuming a minimum orbital period of 12 years, this corresponds to a period-averaged dust production rate of $\lesssim$ 5.5 $\times$ 10$^{-7}$ $M_{\odot}$ yr$^{-1}$. These observations suggest that even a single WC system can contribute to the dust budget at subsolar metallicities, and that such systems are an important source of carbonaceous dust grains in the early universe.
Read the full abstract

All authors Sam Rose, Ryan M. Lau, Jacob E. Jencson, Mansi M. Kasliwal, Andreas A.C. Sander, Howard E. Bond, Michael F. Corcoran, Izumi Endo, Matthew J. Hankins, Olivia C. Jones, Viraj R. Karambelkar, Astrid Lamberts, Thomas Madura, Anthony F.J. Moffat, Patrick W. Morris, Takashi Onaka, Michael E. Ressler, Noel D. Richardson, Christopher M. P. Russell, Itsuki Sakon, J. Sanchez-Bermudez, Gerd Weigelt, Peredur M. Williams

Computer Vision

Less Is More: Balancing Positive and Negative Space in Visual Concept Blending

Graphic designers often blend visual concepts to communicate multiple ideas within a single image, leveraging positive and negative space to create balance, emphasis, and aesthetic appeal. While computational methods have begun to support automatic concept blending, they largely overlook the role of spatial composition in the design. To address this gap, we present an automatic pipeline that explicitly applies positive and negative space throughout the blending process. Our approach first identifies plausible regions for concept integration by combining semantic reasoning from vision-language models with geometric constraints derived from real-world examples. Conditioned on these regions, the system generates blended compositions using a hybrid pixel-vector pipeline: diffusion-based inpainting produces a fast, coarse initialization, which is then refined through vector-based optimization at the point level to ensure structural coherence and balanced semantic expression. A multimodal agent orchestrates this process as a planner and evaluator, enabling iterative improvement and interpretable control. Through an evaluation using both baseline comparisons and a user study, we demonstrate greater expressiveness, creativity, and concept recognizability by effectively leveraging positive and negative space. We further demonstrate the generalizability of our approach across diverse applications, including controllable image and infographic generation.
Read the full abstract

Quantum Physics

Fractal dimension predicts quantum kernel collapse in angle-encoded data

Angle-encoded quantum kernels on tabular data collapse when the feature map is wider than the intrinsic dimension of the data. We propose the correlation fractal dimension D2 as an a priori qubit budget: encode D2 coordinates chosen by FD-ASE instead of the PCA-95% width or all E attributes. On nine data sets and a statevector simulator (n= 32), a one-layer ZZ fidelity kernel at q=D2 stays geometrically alive while the same kernel at the PCA-95% width has already collapsed. The budget is map-dependent: product-state and IQP maps overshoot it; a second ZZ layer undershoots it. Packed dense-angle and re-uploading encodings still live at the fractal q, but not when PCA-95% features are stacked onto those qubits. Shrinking the angle bandwidth moves the ZZ knee later; stretching it kills the kernel earlier. On IBM Quantum (ibm_fez, 256 shots, n=8) the one-layer ZZ kernel at the fractal width matches the exact kernel (MAE 0.021); past that width both hardware and simulator have collapsed. The ceiling is a property of the map-data pair at a stated bandwidth, not of the classical table alone.
Read the full abstract

Computation and Language

Exploring Collaboration between a language and a non-language agent

LLMs are increasingly deployed as orchestrators that coordinate specialized subagents to solve complex tasks through natural language. However, in many important domains like game playing and robotics, the strongest available agents are not language models. Integrating non-language agents with LLMs would require verbalization: compressing their rich continuous representations into sparse textual summaries at each interaction step. To study whether verbalization constitutes a bottleneck, we introduce LLAMIA-Bench, a suite of six diverse collaborative chess tasks spanning three facets: behavioral imitation, state assessment, and natural-language explanation. Each task instantiates a well-established chess problem that neither the LLM nor the chess engine can solve alone. To solve LLM collaboration with non-language agents, we introduce latent state internalization, which projects the subagent's continuous representations directly into the LLM's token stream as learned state tokens, with dynamic re-encoding as actions advance the environment state. Comparing internalization to verbalized integration, our experiments reveal a consistent verbalization debt: the performance gap widens throughout training and persists as the LLM scales from 4B to 14B parameters. A single 14B model, LLAMIA, trained with latent state internalization, matches or exceeds task specialists and frontier models including GPT-5.1 with tool access across all benchmark tasks, and generalizes out-of-distribution where task-specific finetunes collapse
Read the full abstract

All authors Harini S I, Somesh Singh, Yaman K Singla, Rajiv Ratn Shah, David Doermann, Balaji Krishnamurthy

Image and Video Processing

Expert-like Bone Ultrasound Segmentation through Expert-in-the-loop Mask-conditioned Progressive Learning

Manual annotation remains a major bottleneck in ultrasound (US) bone segmentation, where experts typically iteratively refine rough brush masks rather than delineating precise contours in a single pass. We present ExiL, a mask-conditioned progressive learning framework that models annotation as a structured refinement trajectory. ExiL combines a synthetic expert-like brush simulator based on signed distance fields with a lightweight 7.8M-parameter U-Net that learns to complete and refine imperfect masks from US images. During deployment, an expert mode updates the model directly from accepted refinements, enabling continual adaptation to expert behavior. Evaluated using UltraBones100k cadaver data for quantitative segmentation and a prospective volunteer dataset for annotation-efficiency analysis, ExiL reduced single-expert average annotation time from 60 to 20 seconds per frame (66.7%) and improved mean Dice by approximately 0.045 over non-progressive training, while achieving 0.87 Dice and 2.7 px boundary error in the best trajectory-aware setting. With 10–50 ms inference, ExiL enables real-time, self-improving annotation for US-guided orthopedic workflows in practical clinical labeling.
Read the full abstract

All authors Arash Tavangar, Larissa K. Chiu, Hamidreza Khodashenas, Gregory K. Berry, Amir Hooshiar

Machine Learning

Higher Structures in Deep Learning

We provide an expository introduction on the importance of higher-arity tensor operations to deep learning. Then, we conduct a novel empirical investigation of higher-arity phenomenon in trained neural networks, introduce a hypergraphical generalization of the multilayer perceptron, and explore connections to evolutionary algorithms. We conclude with a discussion of promising directions for future research.
Read the full abstract

All authors Michael L. Roberts, Carlos Zapata Carratalá. Nicholas J. Cooper, Lijun Chen, François G. Meyer, Danna Gurari

Optimization and Control

Operational Regimes in Non-Convex Optimization: A Multiplier-Based Taxonomy

This paper introduces a structural taxonomy for constrained non-convex optimization based on the signature of Lagrange multipliers at KKT stationary points. Leveraging a unified game-theoretic interpretation of eight classical algorithm families–including block coordinate descent, ADMM, generalized Benders decomposition, successive convex approximation, interior-point methods, mirror descent, Frank-Wolfe, and Riemannian gradient descent–we show that the normalized multiplier vector carries an algorithm-independent structural fingerprint. Four scale-free shape features of this vector partition the dual space into five operational regimes: Unconstrained, Resource-Limited, Saturation, Strongly-Coupled, and Hybrid. We establish four structural theorems characterizing the partition: invariance under natural KKT symmetries, local stability under data perturbation with explicit Lipschitz margins from Robinson's strong regularity, codimension-one regime transitions, and the topological identification of the Hybrid regime as the Lebesgue-null boundary of the core regimes. A linear-time classifier is proposed with provable guarantees on correctness, iteration stabilization, sample complexity, and online tracking under data drift. Numerical experiments on 104 mixed-integer nonlinear programs and a downlink beamforming instance validate the theoretical predictions. The framework provides a foundational tool for regime-aware algorithm design and robustness analysis in non-convex optimization.
Read the full abstract

Computation and Language

TRIS: A Tri-Layer Retrieval Integrity Sieve Against Knowledge Poisoning

Retrieval-Augmented Generation (RAG) grounds large language models in external corpora, but implicit trust in retrieved documents creates a critical attack surface: PoisonedRAG shows that a handful of crafted passages can dominate dense retrieval and steer generation toward attacker-chosen answers. We present the Tri-Layer Sieve, a middleware defense that sanitizes retrieved evidence through cross-embedding-space clustering with an independent judge model, structural filtering of trigger-payload artifacts, and LLM consistency verification. The design exploits a key weakness of retrieval-stage poisoning: a single document must satisfy one embedding geometry, one internal Trigger-Payload structure, and one generation objective - rarely all three simultaneously, a fragility that persists even against an adaptive attacker who paraphrases around it. On Natural Questions, HotpotQA, and MS-MARCO with Contriever retrieval (k=50), the Sieve reduces black-box Attack Success Rate from 67.0/87.0/64.0% to 3.0/14.0/4.0%, mitigates white-box HotFlip attacks from ~74% to 27.8% on NQ with Layer 3 enabled, and drives poisoned-document MRR to 0.000, while restoring clean accuracy from 13-33% under attack to 58-76%. Under an architecture-aware adversary who paraphrases triggers to evade the structural filter, enabling the consistency layer halves adaptive ASR (32.0% to 15.0% on NQ) while raising clean accuracy by 18 points, at an added latency of ~16-19 s/query under live retrieval.
Read the full abstract

All authors Muhaimin Bin Munir, Akib Jawad Ononto, Nazia Shehnaz Joynab, Bhavani Thuraisingham, Latifur Khan

Computer Vision

SAM3-LoRA: Parameter-Efficient Adaptation of a Concept-Promptable Foundation Model for Multi-Class Structural Defect Segmentation

Promptable segmentation foundation models such as SAM3 accept an open-vocabulary text concept and return every instance matching it, but adapting them to a specialized domain by full fine-tuning is computationally prohibitive for the organizations that would benefit most. This study applies Low-Rank Adaptation (LoRA) to SAM3 for multi-class structural defect segmentation and examines both how such a model can be supervised from conventional annotation and whether the resulting efficiency gain transfers across datasets. Two contributions are methodological. First, we describe a supervision procedure that trains a concept-promptable model directly from COCO-style class-labeled instance segmentation by using the category name itself as the prompt, requiring no prompt templates, no synonym expansion, and no learned class embeddings. Second, we identify and mitigate a failure mode specific to this setting: because a conventional annotation file yields positive prompts exclusively, the model's presence prediction decouples from the text condition and degenerates into responding to any prompt, a collapse that is invisible to every metric computed on positive prompts alone. Exhaustive hard-negative prompting, in which every dataset category absent from an image is issued as a zero-detection query, addresses this at no annotation cost. Two adapter placements were compared under an identical protocol, updating 0.121% and 1.341% of model parameters. On a purpose-built tunnel lining dataset, pixel intersection-over-union improved from 0.017 to 0.338 and instance-level recall from 0.375 to 0.672; on the independent public Structural Defects Dataset, from 0.017 to 0.855 and from 0.574 to 1.000. Improvements were directionally consistent across ten metrics on both datasets, and the largest per-category gains occurred precisely where zero-shot competence was absent.
Read the full abstract

All authors P. Malaisree, S. Youwai, S. Janrungautai, D. Amorndechaphon, P. Rojanavasu, W. Songkitti