Software Engineering
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Sunday 8 November
Pre-training code language models relies on massive code corpora usually streamed over the internet. However, researchers rarely need a multi-terabyte archive in its entirety; instead, their workflow centers on curated subsets: selecting well-documented files, filtering out repetitive boilerplate, auditing for benchmark contamination, or retrieving specific code patterns. Yet, because modern corpora are typically published as file pointers and basic attributes rather than source text, no general structural layer exists to select which files to download: curating a subset requires downloading and parsing millions of files first just to discard most of them. We present SWHForest, an open-source framework and public dataset that provides this missing structural layer for any pointer-based corpus without redistributing source code. A single pass over the archive parses each file into an AST and immediately discards the source text, keeping only its structural summary. Queried locally or exposed by archive providers as a server-side filtering API, SWHForest selects matching files first (via SQL or index intersection) and downloads only surviving candidates. Across the ten largest languages of Stack-Edu (52.1M unique files, 341M definitions), SWHForest answers curation queries in milliseconds while cutting downloads by 46-72%, reducing live structural searches from hours of per-language streaming to seconds or minutes.
Robotics
When Coding Agents Get Arms: Safety Risks Across the Digital-Physical Boundary
Advances in spatial understanding have enabled frontier models to power coding agents that control a real robot arm zero-shot and end-to-end, without task-specific demonstrations or retraining. This brings two new safety challenges. First, the extensive knowledge of frontier models lets these agents perform tasks that earlier VLA and WAM paradigms could not, opening the door to new forms of capability misuse. Second, coding agents can access digital resources such as web pages and emails. Threats in these resources, such as indirect prompt injection, can carry over into the physical world when the agent controls a robot. We therefore introduce ArmedAgent, the first systematic benchmark to evaluate both risks for coding agents performing zero-shot robot control. It covers robustness, malicious use, and indirect injection across 36 tasks in 5 scenarios, evaluated in simulation and validated on a physical robot arm. A distinctive portion of the tasks target harmful behaviors that earlier VLA paradigms could not perform. Testing several frontier models, including GPT-6-Astra, under widely used agent frameworks, we find substantial safety risks in current systems.
Read the full abstract
Computer Vision
How to "train" a surgeon? Benchmarking generalist agents in surgical scene understanding
Evaluating general-purpose multimodal large language models (MLLMs) for surgical scene understanding remains limited due to scarce surgical data and fragmented evaluation protocols. We address these challenges through a two-stage evaluation framework for surgical scene understanding. In Stage I, we introduce SurgKnowBench, a diagnostic benchmark that harmonizes public surgical datasets across procedures and tasks and evaluates proprietary and open-source MLLMs to map the boundary of pretrained surgical competence. SurgKnowBench reveals that current models retain partial visual priors but remain weak on relational and procedural understanding and workflow assessment. Building on this failure frontier and motivated by the scaffolded nature of surgical training, we introduce SurgAgentGym, an agentic environment where general-purpose MLLM agents can use bounded knowledge search over SurgWiki and general-purpose visual grounding for surgical tasks. SurgAgentGym evaluates whether generalist agents can improve surgical understanding at inference time through tool-mediated evidence use, without additional surgical training. Our results show that agentic inference narrows the gap mainly for perception-heavy tasks, while fine-grained localization and expert-style assessment remain persistent bottlenecks. This opens a path toward future studies of agentic surgical understanding.
Read the full abstract
Physics and Society
Retrospective Analysis of Team-Following Trading Profitability in Pumpfun Meme Coins Using Funding Networks
If the right token-creation teams had been known, how much could following them have returned, and what is left when only information available before each launch is used? We study ten separately recorded Pumpfun windows (April 2024 to September 2026), each reconstructed independently from real-SOL reserve states and finite-capital accounts that start with one SOL. Teams are directed funding groups ranked in hindsight. When every team exits at its own best reserve threshold, chosen in hindsight, the main account ends above one SOL in all 8 windows that have a qualifying team (1.10 to 11.32 SOL), but free cash exceeds one SOL in only 5, and marks of positions that never reached their threshold make up 17–94% of equity. If teams are recognised online, from tokens whose first hour has completed, 9 of 19 teams become identifiable only after their last launch; the remaining 10 end above one SOL in 10 cases, with free cash above one SOL in 7; funding links in this step still come from the full window, so even these results are not a fully realisable signal. A large model balance, a result above random cohorts that repeat the threshold search, and a statistical rejection in the selection-matched nulls (3 of 40 tests after Holm adjustment) are three different statements, and these nulls condition on the discovered funding graph, so they do not calibrate the whole selection procedure. Midpoint tests describe a further source of apparent opportunity. A detailed analysis of the best and the worst window, with every analysis parameter varied, shows that the worst result, which is close to break-even, rises several-fold with a stop rule or another funding-graph construction. The results are conditional retrospective model accounts, not executable profits, and not evidence of a causal effect of the SOL price.
Read the full abstract
Computers and Society
From Dialogue to Task: How Cognitive Engagement With Generative Artificial Intelligence Relates to Post-Dialogue Activity and Task Outcomes
Generative AI is increasingly embedded within learning tasks, yet interaction frequency reveals little about cognitive engagement in student-AI dialogue. Little is known about how such engagement is situated within task workflows or how dialogue characteristics and subsequent task activity relate to students' behaviour and task outcomes. This study examined cognitive engagement in student-AI dialogue, its position within task workflows, and associations of dialogue and task-process characteristics with post-dialogue activity and task outcomes. Data were collected from 1,681 students across 40 university courses using three AI companions embedded in resource creation, peer moderation, and practice. Cognitive engagement in 829 conversation threads was coded with a dialogue-oriented Interactive-Constructive-Active-Passive (ICAP) framework, and 2,969 task sequences were modelled with first-order Markov models. Regression models linked dialogue and task-process characteristics to behavioural re-engagement, peer-rated artefact quality, and practice performance. Passive and Active engagement predominated, although resource creation involved more Constructive and Interactive dialogue, peer moderation more Passive, and practice more Active, while higher-engagement dialogue was less often followed by session termination. In resource creation, student-composed messages were associated with more dialogue-to-task returns than template-only requests at the same Passive level, whereas deeper cognitive engagement was associated with higher final artefact quality independent of return frequency; in practice, AI dialogue after an incorrect attempt was associated with fewer further attempts but higher next-attempt correctness among continuers. Post-dialogue activity and task outcomes are distinct aspects of AI use that interaction frequency cannot capture and evaluations should assess separately.
Read the full abstract
Representation Theory
Dlab–Ringel standardization in algebraic triangulated categories
We introduce the notion of derived standardly stratified algebras as a connective dg analogue of standardly stratified algebras. To this end, we extend the Dlab–Ringel standardization method to algebraic triangulated categories and show that every basic proper derived standardly stratified algebra arises from this extended method. We establish Ringel duality for derived standardly stratified algebras using silting objects. More precisely, the Ringel dual is again derived standardly stratified, and the corresponding perfect derived categories are related by mutually quasi-inverse contravariant triangle equivalences. We also show that, over an algebraically closed field, a proper derived standardly stratified algebra is smooth if and only if its derived standard modules are exceptional. Finally, we prove that dg algebras given by finite acyclic graded quivers with arrows of non-positive degrees and homogeneous relations are derived standardly stratified with respect to every total order on the vertices.
Read the full abstract
Computer Vision
Unlearning at Speed: Rapid Activation Dampening for Efficient Machine Unlearning in Real-World Face Recognition
Machine unlearning has become increasingly important in face recognition, where regulations like the GDPR empower users to request the removal of their personal data from deployed models. However, most existing unlearning methods are designed for closed-set classification with small-scale models and training data, and therefore do not generalize to real-world face recognition systems, which operate in an open-set setting with large-scale models and data. In this work, we propose Rapid Activation Dampening (RAD), a fast and targeted unlearning method tailored specifically for large-scale open-set face recognition. RAD approximates differences in average neuron activations between forget and retain identities, identifying and selectively suppressing neurons that are most associated with the targeted identity through a sensitivity-controlled dampening mechanism. We further propose the Forget Score (FS), which addresses the mismatch between existing unlearning metrics and open-set problems by providing a probabilistic interpretation of forgetting directly in the representation space. Experiments across several recent large-scale face recognition architectures and datasets show that RAD achieves strong forgetting while preserving overall model utility and substantially improving unlearning speed. These findings establish RAD as a scalable and practical approach to machine unlearning in face recognition systems.
Read the full abstract
Number Theory
Ranks of fibers in random elliptic surfaces
For random integer polynomials $P$ of arbitrary degree we prove that for $100\%$ of elliptic surfaces $y^2=x^3-P(t)^2x$ the average 2-Selmer rank of the fibers is 3. Consequently, for $100\%$ these random elliptic surfaces we establish a lower bound for the proportion of fibers attaining algebraic rank $0$ and $1$.
Read the full abstract
Probability
Singular parabolic Anderson models - From flat space to geometric contexts
These notes (which grew out of some lecture notes prepared for the first Aarhus Summer School in Analysis and Probability) are an introduction to the parabolic Anderson model (PAM), or otherwise stated the stochastic heat equation $\partial_{t} u \,=\, \tfrac{1}{2}\Delta\, u \,+\, \beta\, u\,\dot{W}$. In this system, a multiplicative random potential $\dot{W}$ drives the solution away from the smooth spreading of the classical heat flow and toward intermittency. Namely one observes that mass concentrates, as time grows, on a few tall and widely separated peaks. Through the Feynman-Kac representation this analytic phenomenon has an equivalent probabilistic reading as the localization of a directed polymer in a random environment. We first develop this picture on the flat space $\mathbb{R}^{d}$, covering the construction of the driving noise, well-posedness of the equation, and two-sided bounds on the moment Lyapunov exponents that quantify intermittency. We then turn to the more original goal of the monograph: understanding how these phenomena, and the exponents attached to them, change when $\mathbb{R}^{d}$ is replaced by a genuinely geometric background, such as a sub-Riemannian manifold or a fractal set. Throughout, the emphasis is on concepts, main results, and the intuition behind their proofs, rather than on exhaustive technical detail.
Read the full abstract
Atomic Physics
High-sensitivity sensing technology of 10 MHz radio frequency electric field based on Rydberg atoms
To address the pressing need for high-sensitivity, miniaturized electric field sensors in the short-wave communication band (3-30 MHz), this study proposes and demonstrates a radio frequency electric field quantum sensing scheme based on cesium atomic Rydberg states and the electromagnetically induced transparency (EIT) effect. The scheme employs an all-infrared optical design, utilizing 852 nm probe light, 1470 nm dressing light, and 780 nm coupling light to efficiently prepare the 49P3/2 Rydberg state in a room-temperature cesium vapor cell, which is then probed non-destructively via ladder-type four-level EIT spectroscopy.
Read the full abstract
Analysis of PDEs
Sharp coercivity estimates and stability loss for the Ginzburg-Landau energy
In this paper we investigate how the coercivity of the Hessian of the Ginzburg-Landau energy at local minimizers $u$ depends on the Ginzburg-Landau parameter $\kappa$ in the regime of large $\kappa$. This question is motivated by its crucial role in the numerical approximation of energy minimizers. Accounting for the degeneracy that is induced by phase invariance and possible additional neutral directions, we establish a universal lower bound of order $\kappa^2$ for the inverse coercivity constant $\rho_u(\kappa)$ of local minimizers with uniformly bounded Hessian kernel dimension. For nonvanishing states that satisfy a quantitative subcriticality condition, we prove a matching upper bound, showing that the quadratic scaling is sharp. On a fixed annulus with a fixed smooth magnetic potential, we then construct nonvanishing local minimizers for which $\rho_u(\kappa)\simeq\kappa^4$. Moreover, we exhibit branches of strictly stable local minimizers along which the inverse coercivity constant diverges as stability is lost. Consequently, no locally bounded upper estimate depending only on $\kappa$ can hold uniformly over all phase-nondegenerate local minimizers, even for a fixed domain and magnetic potential.
Read the full abstract
Machine Learning
CHANNELS4WORLD: An Environment-Scale Dataset for Wireless Foundation Models
Wireless foundation models learn reusable representations across tasks, configurations, and deployment environments. Their pretraining requires environmental breadth, yet wireless datasets often sample few sites densely. We introduce Channels4World, a scene-centered dataset of 100,000 distinct outdoor RF scenes and a procedural method for generating them. Each scene links a generated environment and RF material assignments to 32 base-station (BS) positions, 16,384 UE positions, and aligned variable-length multipath and ordered interaction records at 3.5 and 7~GHz. The corpus comprises 52.43 billion physical BS-UE pairs, 104.86 billion frequency-specific evaluations, and a 29.14-TB packaged payload. Scene identity aligns geometry, placements, channels, and interactions across frequencies, allowing environmental breadth and within-scene observation density to vary independently. The public release offers predefined subsets at multiple scales and separate access to corpus components. A deterministic 90K/10K scene-level train-test split keeps both frequencies and all scene components together. We characterize morphology, RF materials, and transceiver placement across three generation profiles. Controlled propagation and path-retention analyses show that diffraction materially affects retained-link availability and multipath structure, while the retention policy reduces storage substantially with little loss in represented aggregate power. Under matched training exposure, models trained on broader scene collections perform better on held-out scenes, although an equal-episode control shows that scene count alone cannot explain the gain. By making environmental breadth explicit and controllable, Channels4World supports the development and evaluation of wireless foundation models for generalization across deployment environments.
Read the full abstract
Computer Vision
Flex-DDCM: Zero-Shot Task-Oriented Image Compression via Diffusion Trajectory Guidance
Recent zero-shot image compression methods exploit the powerful priors of pretrained diffusion models to produce perceptually pleasing reconstructions without compression-specific training. However, image compression serves not only human viewing but also machine analysis, while existing zero-shot codecs have limited flexibility in meeting the diverse reconstruction requirements of these applications. In this paper, we propose Flex-DDCM, a framework for zero-shot task-oriented image compression via diffusion trajectory guidance. Building on Turbo-DDCM, we adapt the coded diffusion trajectory through task-guided noise selection, embedding the effects of task guidance into the compressed representation. Guidance terms can be applied individually or combined, with timestep-dependent strengths controlling their influence throughout denoising. This design enables task adaptation without introducing additional decoding complexity. Experiments demonstrate flexible trade-offs among pixel fidelity, perceptual quality, and downstream task performance, with competitive performance against trained task-oriented codecs and faster decoding than Diff-ICMH.
Read the full abstract
Computation and Language
Vietnamese Legal Ontology: Every legal answer, traced to the article that grounds it
A legal answer is only as good as the article it rests on. Vietnam's largest open legal corpus pairs 109,543 published questions and answers with 573,851 legal documents, yet its citations are free text: "Khoan 3 Dieu 101 cua Bo luat nay" ("Clause 3, Article 101 of this Code") names neither the code nor its version. Following the five stages of NVIDIA Auto Ontology, Discover, Ground, Validate, Expose and Learn, we turn four open datasets into a knowledge graph of 11.5 M nodes and 16.0 M edges in which every node and edge carries its source record and method, with no duplicate keys, dangling edges or schema violations. Of 67,868 cited provisions, 61,487 resolve to a dated law version, 51,694 to a parsed article, clause or point, and 19,012 are confirmed word for word. The ontology then shows what flat records hide. 36% of the 78,361 answers whose citations resolve now cite a law that is no longer in force; 2,525 verified citations point to an article number that says something else in the current law; 2,520 citations were re-versioned because their words match a different version than their date suggests; and 5,575 document numbers are shared by several local authorities, one of them by 72. Every number in this paper is computed from the graph.
Read the full abstract
Materials Science
Orbital texture of valence band and interband transitions of the type-II multiferroic van der Waals NiI$_2$
NiI$_2$ is a van der Waals semiconductor characterized by complex electronic and magnetic phases, holding promise towards functional applications. For this purpose, a detailed understanding of its band structure and its excitations is essential to elucidate the electronic and transport properties. We present a comprehensive description of the band structure and of the lowest valence-to-conduction band transition in bulk NiI$_2$, by means of time- and angle-resolved photoemission spectroscopy supported by first-principles calculations. We describe the orbital character of Ni and I manifolds, and identify the main momentum-resolved features of the valence band and the shallow minima of the conduction band. With this study, we provide a foundational understanding of the electronic structure of NiI$_2$ and its single-particle interband electronic excitation, a benchmark for theoretical models, and a framework for future investigations on this material.
Read the full abstract
Game Theory
The Query Complexity of Envy-Free Division: Cake, Chores, and Mixed Resources
We study the query complexity of envy-free division for cake, chores, and resources with both positive and negative values. For arbitrary nonnegative nonatomic chore costs, we give a deterministic exact protocol using $2^{n\log_2 n+O(n)}$ queries and cuts. For both cake and chores, a Las Vegas protocol uses $O(n^{12}\log(2n))$ expected queries and $O(n^{11}\log(2n))$ expected cuts. Conversely, every exact protocol with fixed positive success probability requires $\Omega(n^2)$ expected queries in general, even with arbitrarily many output cuts at no query cost. For mixed cake and chores, no uniform finite expected-query bound exists even when every Cut returns the leftmost root or every Cut returns the rightmost root. This holds for any fixed positive multiplicative approximation factor and success probability, already when all net totals are one. An additive impossibility holds on the same normalized domain.
Read the full abstract
Machine Learning
Computationally Efficient Offline Policy Optimization with All-Policy Linear Realizability
We study offline policy optimization in finite-horizon Markov decision processes where every deterministic Markov policy has an action-value function linear in known features, with coefficient norm at most $B$. From independent complete trajectories of one unknown Markov behavior policy, and under all-policy occupancy coverage, we give an algorithm with $\widetilde O(\varepsilon^{-2})$ sample dependence and computational complexity polynomial in $\log(2+B)$. The algorithm performs exactly one convex regression per layer. Its stopped-return targets are the values of a single consistent Markov witness policy. A shared reference design and bounded prediction constraints convert fast clipped-regression error into pointwise confidence for the raw linear predictions. Truncated leverage controls the performance loss without requiring empirical covariance approximation at scale $B^{-2}$. Rounding confidence radii upward preserves the exact witness and yields a polynomial-time bit implementation for rational numerical inputs. The guarantees allow any fixed initial-state distribution; realizability and coverage need only be checked for deterministic policies. We also establish a covariance-coverage variant with uniform approximate realizability and polynomial dependence on $B$. A two-action lower bound shows that the accuracy exponent $2$ is necessary.
Read the full abstract
High Energy Physics (Phenomenology)
How to isolate charm production mechanisms using charm-anticharm rapidity and azimuthal angle differences
Charm production in hadronic collisions proceeds via many different processes, which makes it hard to calculate from first principles. Using PYTHIA simulations, we study correlations between charm and anticharm hadrons and show that hard-process charm, final-state splitting charm, and initial-state-shower charm populate markedly different regions of rapidity and azimuthal-angle difference. In particular, selections on the pair rapidity separation, $\Delta y$, can strongly enrich or suppress the near-side final-state splitting contribution and thereby provide experimental sensitivity to the different production processes. A comparison with existing LHCb $D^0\bar{D}^0$ correlation measurements from Run 1 motivates more differential studies using the substantially larger LHC data sets now available. Extending such measurements to heavy-ion collisions could further distinguish medium modifications of charm produced directly in the hard scattering from charm originating from a parent gluon, providing a new handle on the different interactions of heavy quarks and gluons with the quark-gluon plasma.
Read the full abstract
Computer Vision
GenRM-Flow: Generators are Process-aware Reward Models in Flow Matching
Aligning video generation models with human preference depends critically on the quality of the reward model that scores generated videos. Current video reward models, however, source their reward signal from a model's representational rather than generative capacity: they either repurpose vision-language models as pixel-space scorers, or train auxiliary scoring heads on top of frozen generator features. The generator's actual ability to predict velocities, denoise latents, and produce videos plays no role in scoring its own generations. We argue that this overlooks a structural property of offline preference learning. Any preference optimization objective that approximates an offline preference dataset mathematically targets a strictly defined KL-constrained optimal policy. By isolating the reward term from this analytical solution and mapping the exact generation log-likelihood to its flow-matching surrogate, the empirical reward emerges deterministically as the relative difference in continuous-time velocity-prediction errors. We formalize this equivalence into GenRM-Flow, a framework where a preference-tuned video generator inherently operates as a process-aware reward model directly in the noisy latent space. We further show theoretically that distinct fine-tuning objectives, whether contrastive like DPO and IPO or non-contrastive like RWR, are alternative optimization pathways toward the same optimal policy, each inducing a reward readout from the generator's velocity-prediction errors. Furthermore, the extracted process reward seamlessly substitutes external VLM scorers in FlowAWR. Operating purely on native latent outputs without auxiliary reward networks, this generative reward drives measurable improvements in downstream video generation.
Read the full abstract
Quantum Physics
Universal robust nonadiabatic geometric quantum computation with shortened composite pulses on superconducting circuits
The nonadiabatic geometric quantum computation (NGQC) using the nonadiabatic geometric phases is a promising strategy to realize the high-fidelity quantum operation for large-scale quantum processing. However, conventional NQGC schemes combining with optimal control techniques to further improve the gate fidelity and robustness usually either comes at the cost of prolonged gate time or require complex pulse control and only are limited to several certain specific quantum gates or/and for a certain type error source. Thus, implementing universal and robust geometric gates with simple control while further suppressing gate times remains a major challenge. Here, we propose a scheme for universal robust geometric gates with shortened composite pulses, where the pulse control is simple and enough available evolution parameters can be utilized to effectively reduce the sensitivity to universal systematic errors. Numerical simulations demonstrate that our universal geometric gates set exhibit much stronger error resilience than their conventional single-loop counterparts. Finally, we also verify the robustness advantage of our scheme on realistic superconducting quantum circuits with experimentally demonstrated technology. Therefore, our work paves an alternative way to achieve high-fidelity and robust geometric quantum computation against multiple errors.
Read the full abstract
Optimization and Control
Solution Size Growth as an Obstacle to Iterative Refinement for Linear Optimization
Iterative refinement is a technique used to exponentially boost the precision of the solution to certain problems by solving a sequence of related problem instances. We study the size of the solutions arising from the sequence of problems produced by the iterative refinement algorithm when applied to linear optimization problems (LPs). We introduce a class of problems for which the size of such solutions can grow exponentially with the number of refinement iterations, and we show evidence that to avoid this undesirable behavior, a black-box LP solver would have to be lucky—a concept introduced formally in this paper, by which we mean that the solver would have to return a solution of better quality than what is required by its termination tolerances. Furthermore, we show that the exponential scaling occurs if one uses a natural instantiation of the simplex algorithm using Bland's pivoting rule. We discuss the implications of our lower bound on the solution size for optimization algorithms whose running time depends polynomially on the solution size, such as quantum multiplicative weights update algorithms and quantum matrix games algorithms.
Read the full abstract
Number Theory
Monotonicity of Erdős sums
For $k \geq 1$ let $f_k = \sum_{\Omega(n)=k} 1/(n\log n)$, where $\Omega$ counts prime factors with multiplicity, and let $g_k$ be the same sum restricted to odd $n$. Lichtman showed that $f_k \to 1$, with a global minimum at $k=6$, and Gorodetsky, Lichtman and Wong that $f_k = 1 - 2^{-k}(a_1 k^2 + O(k \log(k+1)))$. We prove that $f_k < f_{k+1}$ for every $k \geq 6$, and that $g_k > g_{k+1}$ for every $k \geq 1$. The two statements come from one analytic distinction. Writing $\sum_k P_k(s) z^k = \zeta(s)^z H(s,z)$, at $s=1$ the point $z=2$ lies on the first singular circle of $H$, and strictly inside the first singular circle of the product with the prime $2$ removed. We show that this distinction decides the direction in general. If the complement of a set of primes $\mathcal{P}$ is thin and its least element is at least $3$, the corresponding sums eventually decrease, so the second theorem is one case of a general phenomenon. The inequalities used at the last step of the first theorem are certified by interval arithmetic.
Read the full abstract
Operator Algebras
Differences Between Quantum Symmetries of Graphs and Graph $C^*$-algebras
We establish several characteristic differences between the quantum symmetry groups of directed graphs and their associated graph $C^*$-algebras in the frameworks of Banica and Joardar-Mandal, respectively, considering the family of finite directed graphs without isolated vertices and without multiple edges.
Read the full abstract
Computation and Language
Modeling individual and item differences in atypicality inferences with ACT-R
Humans engage in pragmatic reasoning when encountering mentions of highly predictable events. Such mentions may appear redundant, leading comprehenders to infer that the event is atypical, and accommodate this inference with an explanation for why it is atypical. However, these inferences are not drawn uniformly across individuals, and experimental items vary in inference rates due to differences in event typicality. We show that these patterns can be predicted with a resource-rational cognitive model in which comprehenders select interpretation strategies based on individual exploration tendencies and item-level predispositions. These parameters are estimated independently from participants' Raven's Matrices score, event typicality rating and language model based accommodation likelihood. Taken together with recent findings from reference games, our results motivate a new perspective on the mechanisms that support pragmatic inferences.
Read the full abstract
Probability
Fourth-Moment Theorems for Sums of Multiple Wigner Integrals
We derive quantitative free fourth-moment bounds for sums of two Wigner chaoses of opposite parity under mirror symmetry. The bound contains an explicit remainder consisting of internal contractions of the two kernels. In particular, vanishing of this remainder together with the fourth free cumulant implies convergence to the standard semicircular law. If the higher-order kernel is cyclically symmetric, the internal contractions can be moved to an endpoint and we obtain an explicit fourth-root bound without a remainder. For fully symmetric kernels we obtain a square-root bound under the $W_2$-Wasserstein distance. Using the operator-norm support estimate for Wigner integrals, we also extend these bounds to every $W_p$-Wasserstein distance , where $1\le p<\infty$.
Read the full abstract
Combinatorics
The strong cylinder conjecture and Rédei's factorization conjecture in three dimensions
In this paper, we prove the strong cylinder conjecture, that is, for every prime $p$, a set of $p^2$ points in $\mathbb F_p^3$ that meets every affine plane in a multiple of $p$ points is a union of $p$ distinct parallel affine lines. We use this result to resolve Rédei's 1970 factorization conjecture in three dimensions: if $A,B\subseteq\mathbb F_p^3$ contain zero and every vector in $\mathbb F_p^3$ has a unique expression $a+b$ with $a\in A$ and $b\in B$, then one of the two sets is contained in a plane through zero.
Read the full abstract
Number Theory
2-Selmer in arithmetic progressions
We prove uniform asymptotics for the 2-Selmer cardinality in the congruent number family $y^2=x^3-D^2x$ with twisting parameter $D$ restricted to long arithmetic progressions.
Information Theory
Joint Design of Communication and Control in Center-Fed Pinching-Antenna Systems
An analytical and optimization framework for the joint design of communication and control (JDCC) in a center-fed pinching-antenna system (C-PASS) is proposed. Within this framework, the mathematical coupling between data throughput and closed-loop control cost is characterized for JDCC. The analysis captures the mutual interference between concurrently transmitted data and control streams during uplink state uploading and downlink command delivery. Based on this coupling, an achievable communication-control performance region is established for C-PASS. Accordingly, a throughput maximization problem is formulated to jointly optimize digital and pinching beamforming subject to control-cost constraints. To address this problem, an alternating optimization algorithm is developed based on constraint reformulation and resource decomposition. The proposed framework reveals how control requirements shape achievable data throughput, guiding the joint beamforming design. Numerical results demonstrate that 1) the proposed C-PASS design achieves 110.2% higher frame-average throughput than conventional PASS, and 2) the proposed concurrent data-control transmission protocol increases frame-average throughput by up to 66.7% over separated transmission.
Read the full abstract
Atomic Physics
Offline Evaluation of Information-Theoretic Adaptive Bayesian Vector Magnetometry
An information-theoretic adaptive Bayesian framework for jointly estimating the longitudinal and transverse components of a magnetic field from the time-resolved optical response of a cold $^{87}$Rb ensemble is presented. A linearly polarized continuous-wave probe drives the $D_2$ $F{=}1 \rightarrow F'{=}0$ transition, producing coherent population trapping (CPT) lineshapes whose width, contrast, and transient dynamics depend sensitively on the ambient magnetic field. The full temporal signal, including early-time transient signal and the steady-state absorption, is compared against a precomputed Maxwell-Bloch forward model. The joint posterior distributions of the longitudinal and transverse components of the magnetic field are updated sequentially, while marginalizing over instrumental nuisance parameters arising from the finite resolution of the magnetic bias coil supply. At each step, the applied longitudinal bias field is chosen to maximize the expected Kullback-Leibler (KL) divergence between the current prior and the anticipated posterior, the expected information gain. As a proof of principle, the protocol is validated in post-processing on pre-recorded experimental scans, and both field components converge to well localized field estimates with weak correlation between them, consistent with the distinct dependence of the CPT resonance position and lineshape on the field components. These results establish an information-theoretic framework for joint vector field estimation from transient cold-atom spectroscopy. However, there exists a systematic forward-model discrepancy, showing that posterior concentration alone does not establish accuracy and motivating nuisance-parameter inference in future online implementations.
Read the full abstract
Analysis of PDEs
Elliptic $L^p$ Carleman estimates near the boundary, quantitative unique continuation, and applications to controllability and stabilization
This article deals with quantitative unique continuation results up to and from the boundary for second-order elliptic equations with variable coefficients via $L^p$ Carleman estimates. The coefficients in the leading part are assumed to be Lipschitz continuous, and the potentials in the lower order term are allowed to belong to almost scale-invariant $L^q$ spaces. As applications, following the method of Lebeau and Robbiano, we obtain a spectral estimate which yields the interior null controllability of the associated parabolic equation, with an explicit dependence of the control cost on the potential. We also prove the logarithmic decay of the energy for damped wave equations with unbounded dampings and potentials.