Peer Revenue Publish or profit

Papers

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

All of arXiv

Monday 7 September

Numerical Analysis

A Generalized Finite Difference-Based Fragile Points Method for Heat Conduction Problems in Non-Homogeneous Media

This paper presents an enhanced formulation of the Fragile Points Method (FPM), a truly meshless approach for efficiently modeling implicit interfaces in two-dimensional differential equations involving non-homogeneous materials. The proposed framework eliminates the need for specialized numerical integration techniques and provides a systematic mathematical foundation for solving interface problems. Discontinuities in both primary and secondary variables across interfaces are naturally handled through the inherently discontinuous shape functions of FPM. Unlike conventional Galerkin methods, FPM employs simple, local, point-based polynomial trial and test functions constructed via a generalized finite difference approach. These discontinuous functions bypass the continuity requirements of standard Galerkin frameworks. To address the resulting inconsistency due to discontinuities, we incorporate numerical flux corrections inspired by the discontinuous Galerkin method. The proposed method is validated through several benchmark problems, demonstrating its efficiency and robustness.
Read the full abstract

High Energy Physics (Phenomenology)

High-Energy Nuclear Recoils from Boosted Dark Matter for the LZ 248-keV Event: Beyond the Halo-Dependent High-Velocity Tail

The LUX-ZEPLIN (LZ) Collaboration has reported a nuclear-recoil candidate at $E_R=248\pm23_{\rm stat}\pm23_{\rm sys}$ keV, with a maximum local significance of $3.4\sigma$ and a global significance of $2.6\sigma$. A prominent interpretation invokes heavy halo dark matter near an inelastic threshold and therefore depends sensitively on the poorly constrained high-speed tail of the Galactic velocity distribution. In this Letter, we propose a qualitatively different possibility based on light boosted dark matter (BDM), whose incident energy is determined primarily by the dark-sector mass spectrum. We consider multi-component scenarios in which the boosted state scatters elastically or inelastically off xenon nuclei. For elastic scattering, pseudoscalar-mediated momentum dependence suppresses low-energy recoils. Near-threshold endothermic scattering of a nearly monochromatic BDM flux can instead confine the signal between kinematically determined recoil endpoints, suppressing events in both the low- and high-energy sidebands. The upscattered state may furthermore decay invisibly within the dark sector, preserving a single-nuclear-recoil signature without requiring it to be detector-stable. We present representative benchmark spectra and discuss complementary tests using other target nuclei and large-volume liquid-scintillator experiments.
Read the full abstract

All authors Haider Alhazmi, Doojin Kim, Kyoungchul Kong, Jong-Chul Park, Seodong Shin

History and Overview of Mathematics

Navigating the AI crisis: a humble guide for students

We present some basic recommendations for (mostly graduate) students when it comes to choosing to use, or avoid, AI in their mathematical research and education. These include: maintaining a connection to the tried and true ways for learning mathematics; focusing on communication and the sharing of ideas with other people; thinking critically about the politics of labor and organizing with your cohort and others at your home institution; and being in touch with the realities of the impact of AI infrastructure on land and people.
Read the full abstract

Robotics

Predictive-Coding-Based Autonomous Regulation of Internally Generated and Externally Coupled Processing in Human-Robot Interaction

Predictive coding characterizes adaptive behavior as a dynamic balance between internally generated predictions and external sensory evidence, yet how an embodied cognitive system can regulate this balance online during ongoing interaction remains poorly understood. This study proposes a predictive-coding-based mechanism for regulating internally generated and externally coupled processing during physical human–robot interaction. The framework employs a predictive-coding-inspired variational recurrent neural network (PV-RNN), in which a meta-prior controls the degree to which posterior inference is constrained by learned prior dynamics. We extend this architecture with an online mechanism that uses reconstruction error accumulated over recent interaction history to select between predefined meta-prior regimes. The mechanism was evaluated across three physical human–robot interaction tasks involving fixed structured, changing structured, and less-constrained interaction. Across all tasks, lower meta-prior values produced the expected increase in posterior–prior divergence and reduction in reconstruction error. More importantly, reconstruction-history-driven regime selection was also associated with reduced prospective prediction error and robot-side physical interaction conflict, demonstrating consequences beyond the retrospective reconstruction objective itself. Task~3 further showed that recent sensory observations can be successfully accommodated while subsequent human motion still departs from the model's prior-generated future trajectory. Overall, these findings show that accumulated reconstruction mismatch can provide an endogenous signal for regulating how strongly subsequent inference relies on learned internal dynamics relative to ongoing sensory input during embodied interaction.
Read the full abstract

Materials Science

Human-agent discovery of reconfigurable in-plane ferroelectric superdomain control

Automated experimentation is most effective when the observables, available actions, and objective are defined before the experiment starts, as is the case for Bayesian optimization. However, in many exploratory experiments, the variables that describe the sample must be extracted from the data, new operations emerge during the experiments, and the instrument budget is too small to learn the problem by trials. Here we introduce the Scanning Probe Agentic Research Cycle (SPARC) framework, in which a coding agent and a human operator share one microscope, one notebook, and two persistent memory files. FINDINGS.md stores graded conclusions about the experiment, whereas PITFALLS.md records learned failure modes of analysis and instrument. We apply SPARC to reconfigure the in-plane superdomain direction of a (111)-oriented PbZr0.2Ti0.8O3 film. In an operator-supervised campaign, the agent reanalyzed earlier manual measurements and developed an oriented lattice of stationary bias pulses with alternating polarity to reconfigure the superdomain direction. In a subsequent agent-controlled campaign, PITFALLS.md entries were compiled into checks that validate a design before any write. The experiments showed that spatial polarity alternation, instead of the exact matching between the lattice and lamellar periods, determines directional selection. Combining a raster scan with a masked pulse lattice printed the letters UTK into the superdomain orientation. The campaign also identified practical requirements for agentic experimentation where physical verification of instrument execution, the conditions under which stored findings remain valid, validation of new observables on instrument data, and robust control protocols.
Read the full abstract

All authors Yu Liu, Boris Slautin, Ching-Che Lin, Jaegyu Kim, Lane W. Martin, Sergei V. Kalinin

Statistical Methodology

Finite-Boundary Reduction and Exact Verification of Strong Familywise Error in Active-Count-Coupled Multi-Arm Efficacy-Toxicity Monitoring

Randomized dose-optimization trials may screen several candidate doses using binary efficacy and toxicity outcomes. A dose is inadmissible if efficacy is insufficient or toxicity is excessive, so each dose-specific null hypothesis is a union region and strong familywise error control must hold across arbitrary mixtures of inadmissible and promising doses. We study exact verification of multistage monitoring rules in which armwise decisions may be coupled through the number of active arms remaining at each interim analysis. For uncoupled rules, we derive an exact product representation and show that a complete-null boundary configuration is least favourable. Under active-count coupling, this factorization no longer holds because a promising dose can remain active and alter future boundaries for null doses. Assuming independent sampling across arms, prespecified per-dose analysis schedules, analysis timing not driven by the observed efficacy or toxicity outcomes, arm-specific monitoring statistics, and stagewise monotonicity, we establish an exact finite-boundary characterization without parametric restrictions on the within-arm joint efficacy-toxicity distribution. Each dose need only be evaluated at an efficacy-null boundary, a toxicity-null boundary, or a maximally favourable alternative. A deterministic finite-state recursion then verifies a fixed decision table without Monte Carlo error. Numerical studies confirmed the prespecified strong familywise error control and showed that active-count coupling can shift the least-favourable configuration away from the complete null while improving joint retention of multiple promising doses. A published randomized dose-ranging trial was used as a clinical illustration of how the framework could be prospectively implemented. The framework separates monitoring-rule construction from rigorous error verification.
Read the full abstract

Earth and Planetary Astrophysics

Constraining the orbit of the retrograde planet in the $\nu$ Octantis system

The $\nu$ Octantis system is composed of a compact stellar binary and a retrograde S-type planet orbiting near the edge of stability. Here, we investigate the dynamically viable architectures of this system by exploring the planetary orbital elements with a Monte Carlo method. To handle the large number of free parameters, we employed a novel approach in which all relevant orbital parameters were varied simultaneously, which we refer to as corner stability maps. Furthermore, since the osculating orbital period of the planet undergoes large oscillations, we combined these maps with frequency analysis to identify the stable configurations that remain consistent with the observational data. We find that the retrograde planetary orbit most likely lies in the region of influence of the high-order 28/$-$11 mean-motion resonance, where its long-term stability is enhanced by secular apsidal alignment.
Read the full abstract

All authors Alan Cefali Signor, Alexandre C.M. Correia, Maria Helena Moreira Morais, Timothée Vaillant, Ho Wan Cheng, Man Hoi Lee, Trifon Trifonov

Other Condensed Matter

On average stress within sub-regions of granular media

The paper concerns computation of average stress within small sub-regions of a larger static granular assembly, where the sub-region's boundary is allowed to pass through the assembly's particles. An exact average is computed for certain stress components and for certain categories of sub-regions. The paper also identifies those choices of sub-regions for which exact stress components may be exactly computed, but when these conditions are not met, provides reasonable bounds on the error.
Read the full abstract

Probability

Strong Galerkin Approximation, Malliavin Regularity, and Blow-Up for a Mixed Local–Nonlocal Stochastic Wave Equation

We investigate the dynamical behavior of a class of semilinear stochastic wave equations on a bounded smooth domain $\Ocal\subset\R^d$ driven by additive trace-class noise, where the elastic response is governed by a mixed local–nonlocal operator $\Acal=-\theta\Delta+\beta(-\Delta)^s$ with $s\in(0,1)$. A fundamental challenge in this setting is that the local and nonlocal operators do not commute on bounded domains: the natural Dirichlet basis fails to diagonalize the restricted fractional Laplacian. Consequently, we first establish the strong convergence of the resulting non-diagonal, dense Galerkin approximation scheme. Leveraging these uniform energy bounds, we rigorously derive the associated Itô energy identity. In the defocusing regime ($\varepsilon = +1$), this strong approximation yields global well-posedness on the energy-subcritical range, providing a unique probabilistically strong solution in the energy space $V \times H$. Within this variational framework, we conduct an analysis of the Malliavin regularity, showing $(u(t), v(t)) \in \mathbb{D}^{1,2}(V) \times \mathbb{D}^{1,2}(H)$, and leverage fractional Sobolev embeddings to prove that the one-dimensional probability law of $u(t, x_0)$ is absolutely continuous via the Bouleau–Hirsch criterion. In stark contrast, for the focusing regime ($\varepsilon = -1$), we establish local well-posedness and prove a rigorous dichotomy: under a negativity condition on the initial energy, either pathwise explosion occurs with positive probability in finite time, or the energy norm possesses an infinite second moment before an explicit critical time $T^*$. Finally, we observe how the dense Galerkin interaction matrices pose unique structural challenges for spatial statistical inference.
Read the full abstract

Machine Learning

Noisy-Space Policy Gradient for Diffusion Policies in Offline Reinforcement Learning

Diffusion policies offer a powerful and expressive parameterization for continuous control. Yet, their integration with reinforcement learning remains conceptually and algorithmically challenging. In this work, we address this gap by introducing a noisy-space action-value (Q-)function that assigns values to diffusion latents through the distribution of executed actions induced by the denoising process. We show that this construction admits a precise semantic interpretation and derive a noisy-space policy gradient (NSPG) that optimizes noisy latents using only clean action-space value estimates. Building on this result, we formulate a KL-regularized policy improvement over noisy latents and show that the resulting objective admits a diffusion-compatible regression form, avoiding backpropagation through the denoising process. Empirical results on state-based D4RL benchmarks and vision-based OGBench tasks demonstrate that the proposed noisy-space objective provides a principled and effective basis for training diffusion policies in offline reinforcement learning. Project webpage: https://mahmoud-selim.github.io/NSPG/
Read the full abstract

Machine Learning

Learning Adaptive SED for heterogeneous load balancing

We study a two-server load balancing system with heterogeneous service rates that are a priori unknown to the dispatcher. The goal is to route customers according to the Shortest–Expected–Delay (SED) policy, but this requires knowledge of the service rates. Empirical policies that route based on estimates perform poorly: due to estimation error, the empirical policy disagrees with the oracle on an infinite region of the state space. We propose an online learning algorithm that converges to SED while learning the service rates. The algorithm carefully balances empirical SED routing with forced exploration phases that guarantee sufficient sampling of both servers. We prove that our algorithm achieves finite regret; this differs from classical Multi-Armed Bandit settings where regret typically grows logarithmically in time. Finally, numerical experiments demonstrate the performance of our algorithm and highlight the regimes in which forced exploration is especially beneficial.
Read the full abstract

All authors Sanne van Kempen, Jaron Sanders, Fiona Sloothaak, Maarten G. Wolf

Computer Vision

Contextual Observer Grounding: Evaluating Situated Spatial Reasoning in Vision-Language Models

Reasoning over language instructions in embodied tasks such as robotics often requires understanding spatial relations from a speaker's situated perspective. Humans infer such perspectives from shared environmental knowledge, activity context, and commonsense. Recent vision-language models (VLMs) appear capable of spatial reasoning, but their ability to infer a speaker's viewpoint from contextual cues and interpret situated spatial relations from that viewpoint remains unclear. We call this capability contextual observer grounding. To study this capability, we construct the Point-of-View Benchmark (POVBench), a dataset of 3D scenes and queries that disentangles Inferred, Stated, and Given forms of observer grounding in natural embodied communication. Given multi-view observations and a natural-language sentence, models must localize unseen or underspecified targets from situated spatial and contextual cues. Across state-of-the-art VLMs, localizing targets from directional language remains challenging, even when observer grounding is made explicit. We find that explicit breakdowns of observer-relative spatial reasoning improve target localization. Our project page is available at https://mimo-owl.github.io/POVBench/.
Read the full abstract

Sunday 6 September

Software Engineering

From AIOps Pipelines to Agentic Service Operations: A Survey of Capability, Authority, and Evidence

Modern service systems combine rapidly changing dependencies, heterogeneous telemetry, and operational actions whose side effects may amplify an incident. Conventional AIOps research has produced effective detectors, localizers, and diagnostic models, yet surveys typically organize them by task, telemetry modality, or learning technique. Such taxonomies describe what a method predicts, but not which decisions it may make or which evidence justifies that authority. We survey the transition from AIOps pipelines to agentic service operations through three coupled lenses: capability, operational authority, and evaluation evidence. We first relate SOA, microservices, Kubernetes, service meshes, and serverless platforms to the incident lifecycle. We then review conventional AIOps methods and audit 22 language-model-enabled systems according to their evaluated behavior rather than their intended autonomy. The audit shows that most systems remain copilots or read-only investigators. Only two exercise write authority; both use constrained action envelopes, and neither carries controlled production-outcome evidence. Building on these findings, we synthesize a four-plane reference architecture, evidence grades, an evaluation ladder, and a controlled adoption path. Together, these artifacts clarify which claims current evaluations support and what evidence is still required for dependable agentic service operations.
Read the full abstract

All authors Hailiang Zhao, Wenzhuo Qian, Jianwei Yin, Shuiguang Deng

Computation and Language

Understanding Failures in LLM Reasoning by Learning Structured Representations of Chain-of-Thought

Chain-of-Thought (CoT) is central to improving reasoning in large language models, yet existing evaluation methods focus on step-level signals and ignore the underlying structure of reasoning traces. This raises a key question: what aspects of reasoning quality (i.e., correctness, efficiency, and failure patterns) are encoded in the structure of a CoT trace? We introduce CoT2Graph, a framework that represents CoT traces as directed dependency graphs, making relationships between reasoning steps explicit and capturing patterns such as backtracking, branching, and subtask merging. We model these graphs with graph neural networks and use attribution-based explanations to identify structural patterns associated with reasoning quality. CoT2Graph reveals distinct patterns of successful and failed reasoning: correct CoTs place greater importance on early planning, whereas incorrect CoTs are associated with overloaded synthesis steps and dependency violations. We further show that graph-based signals better predict whether a CoT leads to a correct answer, outperforming previous tree-based structural methods by up to 2.5% and simple length-based baselines by 7.8%, while also improving Best-of-$N$ selection. Overall, our results demonstrate that reasoning structure encodes rich signals about reasoning quality, motivating a shift from sequence-level evaluation to structural analysis.
Read the full abstract

All authors Tommaso Felice Banfi, Mubashara Akhtar, Jingwei Ni, Mrinmaya Sachan

Human-Computer Interaction

Autonomy Is Absent, Not Eroded: An Observational Study of User Autonomy in Multi-Turn LLM Conversations

Does prolonged interaction with Large Language Models reduce users' observable autonomy during conversation? We test this degradation hypothesis by scoring user autonomy across 2,877 multi-turn conversations (46,952 turns) from WildChat and validating on 500 conversations from LMSYS-Chat-1M. Our core finding is a null result: population-level autonomy remains stable throughout conversations, only 4.7% of conversations show meaningful decline, and an apparent decrease at later turn positions is explained by survivorship bias. Instead, the dominant pattern is not erosion but absence: 80.8% of conversations exhibit zero or low autonomy throughout. When autonomy engagement occurs, it is self-reinforcing (persistence doubles from 40% to 85% over consecutive turns). These findings reframe the problem: the risk is not that LLMs progressively degrade users' observable autonomy, but that most users never engage in independent reasoning during the interaction.
Read the full abstract

Human-Computer Interaction

Grounding Insight: An Artificial Affect Signal for Human-Centric Explainability of RL Agents

In Human-Interactive Robot Learning (HIRL), human teachers assess learning progress by observing agent behavior. However, Reinforcement Learning (RL) agents often fail to express learning progress in human-interpretable ways, leading to suboptimal or counterproductive teaching. This work proposes "Insight": a human-interpretable affective signal grounded in the RL process. It is specifically designed to address the "Absence of learning" human teacher mental model misalignment - the misconception that a failing or exploring agent is broken, underpowered or not learning at all - by signaling an agent's "Aha!" Insight moment. The signal is derived from the absolute Temporal Difference (TD) error making it easily implementable in other RL agents. To ground this signal, five design criteria were formulated: Relevance, Interpretability, Faithfulness, Scalability, Generalizability. The structural construct validity was assessed through simulation, examining micro and macro dynamics and event-related sensitivity. Micro dynamics reveal that Insight captures meaningful learning events at the appropriate time and magnitude. Macro dynamics demonstrate that the signal follows the agent's learning phases and reflects overall progress. Event-related sensitivity shows, the signal distinguishes the saliency of environment interactions, such as positive rewards versus "trauma-like" negative events. These results demonstrate that technical RL dynamics can be effectively packaged into human-like affective substrates to close the teacher-learner loop.
Read the full abstract

General Mathematics

Generalized supremum convolutions on the tensor product of unit intervals: t-conorms and triangle functions

The complete-lattice tensor product $\mathbb{E}=[0,1]\otimes[0,1]$ arises naturally in the algebra of truth values for type-2 fuzzy sets. For a monotone surjective map $S$ that is continuous on $[0,1[^2$ and a monotone map $T$ having $1$ as identity, we consider the generalized supremum convolution \[ (f\otimes_{_{S,T}}g)(x)=\sup_{S(u,v)=x}T(f(u),g(v)) \] on $\mathbb{E}$. Rather than assuming classical algebraic structures for $S$ and $T$ from the outset, we ask when they are forced by requiring the induced convolution itself to be a t-conorm. We prove that $\otimes_{_{S,T}}$ is a t-conorm on $\mathbb{E}$ if and only if $S$ is a continuous t-conorm and either $S=\max$ or $T$ is a right-continuous t-norm; in the exceptional case $S=\max$, the convolution is simply the pointwise maximum for every admissible $T$. The same convolution mechanism also yields an order-dual description of triangle functions on the space of distance distribution functions $\Delta^+$. Using an order-reversing isomorphism between $\Delta^+$ and a boundary variant of $\mathbb{E}$, we show that $\tau_{_{T^*,L}}$ is a triangle function precisely when $L$ is a continuous t-conorm with no zero divisor and, unless $L=\max$, $T^*$ is a left-continuous t-conorm; moreover, when $L\neq\max$, continuity of $\tau_{_{T^*,L}}$ is equivalent to continuity of $T^*$. This resolves the remaining $\tau_{_{T^*,L}}$ part of the Schweizer–Sklar problem.
Read the full abstract

Computation and Language

The FlightFinder Occurrence Corpus: harmonising 128 aviation-safety source codes into one deduplicated cross-national accident dataset

National aviation-investigation databases are deep but territorial; global compilations such as the Aviation Safety Network and the Bureau of Aircraft Accidents Archives are hand-curated, apply their own inclusion thresholds, and publish no per-record lineage. The FlightFinder Occurrence Corpus harmonizes 279,485 occurrences and 131,882 attached narratives from 128 source codes – national accident-investigation authorities, occurrence-reporting systems and four global compilations. Records are mapped into one schema; each normalized record reduces to exactly one key from an ordered ladder – event signature, registration and date, rounded position and date, or source and native identifier – so a missing field never creates a link, and the key is the occurrence identifier. Records the ladder cannot reach are scored in a blocked pass into an explicit merge table and applied only within an automatically approved score band or after a person confirms them. Field conflicts follow a documented source ranking; one provenance row per contributing source – its native identifier and URL – is retained, with attribution and licence terms per source code. Linkage error is measured on five adjudicated samples totalling 200 pairs. A recorded uniform random draw of 60 from the 3,441 merges the automatic band applies as it ships found no error; the exact two-sided 95% Clopper-Pearson upper bound on its false-merge rate is 6.0%, or 205 pairs. The measured error covers the scored band only; the ladder's deterministic joins are not adjudicated. The corpus has no exposure denominators and inherits the under-reporting of its sources. It is offered as a base for cross-national safety statistics, source-disagreement studies and narrative text mining; a frozen, DOI-referenced snapshot accompanies the live API.
Read the full abstract

Machine Learning

"very likely" Means "uncertain"? How LLMs Diverge from Humans in Linguistic Uncertainty Quantification

Humans express uncertainty verbally via markers (e.g., "possible," "likely"), yet most LLM uncertainty quantification (UQ) relies on costing likelihood- or consistency-based signals. From a cognitive perspective, accurate verbal uncertainty reflects metacognitive monitoring, representing knowledge boundaries ("knowing that you don't know") to support regulation and information seeking. In this paper, we investigate how LLMs diverge from humans in verbal uncertainty quantification and whether verbal markers can reliably quantify LLM uncertainty. We curate a corpus of human uncertainty markers from psychology and decision-science literature and benchmark LLMs against it. We observe that LLMs encode verbal uncertainty with numerical levels that differ substantially from those of humans. We then introduce METHODNAME, a novel optimization-based algorithm that learns an optimal uncertainty profile over uncertainty markers directly from LLM outputs. By fitting a marker-uncertainty mapping to best explain empirical correctness, METHODNAME discovers how much probability mass each verbal marker should convey, rather than estimating uncertainty via repeated sampling. METHODNAME enables a direct, marker-level comparison of confidence semantics between humans and LLMs, disentangling mismatch and revealing systematic confidence disparities in verbal expressions.
Read the full abstract

All authors Jinhao Duan, Zicheng Liu, Zijie Liu, Kaidi Xu, Tianlong Chen

Fluid Dynamics

A Zero-Flux Material Cylinder in the Unsteady Kiknadze–Krasnov Vortex

In viscous flow, the circulation enclosed by a deterministic material loop generally changes through vorticity diffusion. It is shown that the source-bearing unsteady Kiknadze–Krasnov vortex possesses a distinguished cylindrical material surface on which this viscous circulation transfer vanishes identically. For a single distributed mode, the scaled squared radius $x=\beta(t)r^2$ satisfies \[ \frac{\mathrm D x}{\mathrm D t} = 4\nu\beta(t)\big[(s-1)-x\big], \] where $\beta(t)$ is the inverse squared radial scale, $\nu$ is the kinematic viscosity, and $s$ is the incomplete-gamma shape parameter. For $s>1$, the cylinder $x=s-1$ is transported exactly by the radial flow and preserves the inside/outside ordering of fluid trajectories. The same factor $(s-1)-x$ governs the radial vorticity gradient and the material circulation-transfer rate. Consequently, the magnitude of the distributed axial vorticity $|\omega_z|$ attains its annular maximum on this material cylinder, $\partial_r\omega_z=0$ there, and the enclosed circulation remains exactly $\Gamma P(s,s-1)$ for every admissible prescribed strain history. Neighbouring material cylinders continue to exchange circulation viscously, while the distinguished moving cylinder carries a fixed fraction of the distributed circulation throughout the admissible evolution. In the steady source–strain limit, the surface reduces to the radial stagnation cylinder. The complete velocity and pressure fields, exact radial-scale evolution, admissible parameter ranges, and qualifications for multiple radial scales are also derived. The resulting solution-specific material-circulation invariant provides an exact benchmark for numerical material-surface tracking, vortex-core evolution, and viscous circulation transport, with the principal analytical results additionally supported by machine-checked formal verification.
Read the full abstract

Computational Complexity

A Full Complexity Dichotomy for Complex-Valued Boolean Holant Problems

A Boolean Holant problem evaluates a graph partition function specified by local signatures at its vertices. We prove a complexity dichotomy for arbitrary finite sets of algebraic complex-valued signatures on general graphs: an explicit, decidable criterion characterizes the polynomial-time cases, and every remaining case is $\#\mathsf{P}$-hard under polynomial-time Turing reductions. The proof resolves the all-even case by constructing signatures with nonzero endpoints and all remaining support on the central layer, then classifying them according to half their arity. Binary transfer groups and matching-product reconstruction provide the main tools in the small- and large-height cases, respectively.
Read the full abstract

Quantum Physics

Sharp Target-Domain Certificates for Quantum-Kernel Advantage under Distribution Shift

Quantum predictive advantage under shift usually requires known target labels. We derive the assumption-free sharp identified interval for the finite-batch advantage of a fixed candidate over the best member of a prespecified fixed classical-kernel family under any bounded loss and unrestricted completions of the unaudited labels; zero-one accuracy has exact closed-form partial-label updates. Across eight security shifts, zero-label upper endpoints against 115 classical kernels span 0.002-0.088. Retrospective auditing reduces every endpoint to 0.010 with 0-33 of 500 labels, including entangling-ZZ models. Prospective corroboration on two eligible tasks met predefined criteria: four task-classifier medians require 0-3 labels (overall median 0; maximum 76), and all 20 realized effects are negative; a third task fails its feature gate. Protocol controls reverse an apparent advantage, while finite-shot noise can increase predictive distinctness without useful advantage. The framework separates protocol validity, target predictive indispensability, and quantum relevance.
Read the full abstract

All authors Roberto Fernández-Barrios, Iker Pastor-López, Asier González-Santocildes, Pablo García Bringas

Computational Complexity

When Matchgate Base Collapse Fails: A Qutrit Trichotomy and Unbounded Exact Width

Holographic algorithms solve planar counting problems by encoding each value of a finite domain into several Boolean matchgate wires. Base collapse asks whether every exact representation can be compressed to a number of wires per edge bounded only by the domain size. Chen (STOC 2016) and Xia (STOC 2016) established broad positive collapse theorems under full-rank and related structural hypotheses. Together, these works left open whether a domain-only collapse bound survives when several locally deficient signatures must share one encoding. We resolve this open problem negatively, already on a three-state (qutrit) domain. For every $a\ge4$, we construct a five-label integer-valued qutrit language of maximum arity $a$ with an exact rational matchgate presentation, yet every exactly equivalent label-preserving presentation requires common width $\Omega(2^{a/2}/a)$, even if its domain, base, tensors, and complex weights may all change. The construction avoids the usual local degeneracies, so the obstruction is genuinely simultaneous. Algebraic geometry then drives an exhaustive trichotomy explaining the boundary of collapse: the gadget closure either separates into rays, is Gaussian-mobile and collapses to width at most three, or is confined to a rigid plane-plus-ray flag containing our unbounded family. Pure-spinor geometry forces compression in the mobile case, while dimension bounds for matchgate varieties and Zariski avoidance place exact integer data outside every low-width algebraic image. Thus one geometric framework explains both why collapse occurs and why it can fail without bound.
Read the full abstract

Formal Languages and Automata

Supervisory Control under Partial Observation: Where Observation Consistency Becomes Decidable

Observation consistency (OC) and modified observation consistency (MOC) are structural conditions used in hierarchical and modular supervisory control under partial observation. Their verification for languages generated by deterministic finite automata is PSPACE-hard, whereas decidability was open. We answer this question by showing that both problems are undecidable, that is, there are no algorithms verifying OC or MOC. On the positive side, we identify a decidable class defined by a restriction on the plant: if every cycle of the automaton contains a transition labeled by an observable high-level event, then verification of both conditions is PSPACE-complete.
Read the full abstract

All authors Shaowen Miao, Jan Komenda, Tomáš Masopust, Yiding Ji

Computation and Language

Self-Cleaning and Captured Anyway: One Measured Primitive for Error in a Store an Agent Writes to Itself, and What a Falling Score Actually Measures

"An agent that writes its conclusions into a store it later retrieves from closes a loop usually reported as one-way contamination. Taking the loop to the infinite-tenure limit against an append-only store gives a different picture: because writing never deletes, the reachable state space has a hard upper edge at (n-1)/n, so the outcome is a choice between two edges rather than a decay. At f_0 = 0.9 the interval between the two modes holds 3.6% of 220 runs where a uniform spread would put 20.6%, and is strictly empty on the first 15; the pooled mean describes 8.2% of the runs it summarises, the median 68.2%. Everything the model contributes is carried by one measured primitive with no fitted parameter, the copy function \gamma(\phi): on 36 Wikidata facts, sign(\hat{\gamma} - \gamma_{crit}), with \gamma_{crit} = 1/k at r = 0, w = 1, predicts the direction of drift on 353 of 360 real-fact runs (39 of 40 synthetic in the same batch). Scale does not rescue the store: pooled frontier capture is 0.850, with claude-sonnet-4.5 captured on 20 of 20 seeds against our registered prediction of <0.5. What the interval tests is distinguishability rather than count: on the real facts, multi-valued runs have 6.4x its occupancy of the rest. It survives at f_0 in {0.1, 0.3, 0.5}, capture peaks at f_0 = 0.5, and of four interventions with criteria frozen first, timing dominates fraction at matched budget while a consistency gate drives every model to 0.993. The resampling unit is the seed, at a design effect of 3.75 on a pooled level: under a 44-seed control the ordering supporting claim 4 collapses from Spearman +0.98 at three seeds to +0.31-0.80 at forty-four, while claim 2's ordering is exact there (+1.00, p = 0.017). All 87 graded rows are in Appendix W, 37 of them graded withdrawn, failed, self-correcting, undecidable or an acknowledged limit, against 50 that are not."
Read the full abstract

All authors Wenhui Chen, Jianlin Chen, Ziyao Lin, Chi Man Vong

Computation and Language

LLM-Driven Training-free Location-Attribute Synergic Fusion: A Closed-Loop Paradigm for Dual-source Encrypted POIs and LULC Mapping

Dual-source encrypted points of interest (DSEP), POIs from two encrypted coordinate systems, suffer from intertwined location and attribute uncertainties, including nonlinear systematic misalignment and naming inconsistency, hindering land-use/land-cover (LULC) mapping. To the best of our knowledge, this paper is the first to propose an LLM-driven, training-free location-attribute synergic closed-loop optimization paradigm for DSEP fusion. The paradigm jointly refines location transformation and attribute correspondences through iterative feedback. Attribute-synergic location fusion uses an LLM-driven attribute matching method to establish DSEP correspondences, reducing matching complexity from O(N^2) to O(N), and refines transformation coefficients using an improved particle swarm optimization algorithm within ISODATA-clustered local subregions. Location-synergic attribute fusion then reassesses attribute confidence from updated geometric residuals through an LLM-fuzzy method. The refined correspondences feed back into location optimization, forming a bidirectional closed loop. Sample purification and adaptive radius contraction enable convergence in essentially two iterations. We further propose a training-free LULC mapping method that inherits land-use classes from encrypted maps through location fusion, producing vector-raster integrated LULC maps. A reference-free POI fusion evaluation method is applied across 31 provincial capitals and municipalities in mainland China. Experiments show that our method achieves an average DSEP location fusion residual of 4.58 m and attribute fusion accuracy of 95.12%, improving upon the open-source baseline and state-of-the-art method by 1.77 m and 14.87%, respectively. Overall, the method provides a training-free solution for DSEP fusion and enables georeferencing of encrypted vector data to WGS-84 without field-surveyed ground control points.
Read the full abstract

All authors Chang Li, Xingtao Peng, Yongjun Zhang, Yinfei He, Cairun Huang

Computation and Language

FrontierMath Erdős

We introduce FrontierMath Erdős (FME), a benchmark of 68 Erdős problems that are open as of August 2026. To solve a task in FME, AI systems must resolve (prove or disprove) one of the 68 conjectures in the proof assistant Lean. Our 68 problems were selected by the second author among 652 open problems on erdosproblems.com for their mathematical interest and difficulty. AIs have recently resolved several open problems in mathematics, but these demonstrations fall short of a systematic study of AI capabilities. FME evaluates every AI model on the same fixed problems, autonomously and under the same budget. We evaluated five AIs with a budget of \$300 per problem. One (GPT-6 Astra) scored 3%, and all others scored 0%.
Read the full abstract

Computation and Language

Mitigating LLM Over-Refusal via Dynamic Semantic Routing Calibration

Large language models (LLMs) aligned for safety often suffer from over-refusal, incorrectly rejecting benign yet safety-related instructions. Prior studies primarily attribute this to static representation overlap, largely overlooking the underlying dynamic mechanisms. In this paper, we present the mechanistic analysis of over-refusal through the lens of internal routing conflicts within transformer attention. We discover that a sparse subset of Hypersensitive Safety Heads misfires on Hard-Safe prompts, exhibiting abnormal attention entanglement that forcefully binds harmless target entities to refusal semantics. This triggers a severe, high-entropy routing conflict that deprives target entities of necessary attention. To counteract this, we propose Semantic Routing Calibration (SRC), a lightweight, training-free inference framework. SRC precisely localizes and dynamically suppresses these hypersensitive safety heads at the inference stage. Coupled with a dual-branch logits fusion that acts as a safety regularizer during subsequent decoding, SRC seamlessly restores trustworthy reasoning. Extensive experiments demonstrate that SRC alleviates over-refusal, with intrinsic safety performance preserved as much as feasible.
Read the full abstract

All authors Zixuan Wang, Bingjie Zhang, He Zhao, Dandan Guo

Computation and Language

Teaching a Moving Student: Rethinking the Curriculum of On-Policy Distillation

In on-policy distillation (OPD), the student determines which states receive teacher supervision. As its policy evolves, earlier response prefixes become less likely even though teacher-student disagreement on them persists. Under matched trajectory and optimization budgets, neither more queries nor more frequent rollout resampling is uniformly beneficial. Current-policy rollouts outperform initial-policy rollouts at shorter response budgets, but this ranking reverses at longer budgets. Replaying initial-policy rollouts after current-policy training improves accuracy, whereas replaying fixed recent rollouts does not reproduce the gain. We propose R-OPD, a gradient-triggered curriculum that adaptively selects when to revisit initial-student trajectories. When changes in mean gradients fall within minibatch-level variation for two consecutive comparisons, training switches from the next iteration onward to initial-policy replay. Across eight mathematics benchmarks and three training-data orders, R-OPD improves average accuracy over continued current-policy sampling by 2.25/4.44 percentage points at 16K/32K for a 0.6B student and 2.81/6.15 points for a 1.7B student. With a 30B-A3B teacher, it improves 8B accuracy by 4.05 points at 32K. At 32K, R-OPD exceeds a fixed schedule of 40 current-policy updates followed by 20 replay updates by 1.39/1.66/1.85 points for 0.6B/1.7B/8B. At the same generation cap, R-OPD also produces longer responses, suggesting that well-timed revisits help students use more of their reasoning capacity.
Read the full abstract

All authors Lingxiang Hu, Tianle Xia, Yiding Sun, Ming Xu, Linfang Shang, Lan Xu, Ning Zheng, Wei Xu, Jie Jiang

Computation and Language

AIBuildAI-2.5: Efficient Autonomous AI Model Development Through LLM-Guided Tree Search

Autonomous agents that automatically build artificial intelligence (AI) models could broaden access to AI across science and engineering. A popular line of such agents frames model building as a code search problem and solves it by tree search, in which each node is a candidate program and the tree grows by generating a child program from a parent, and these agents now approach the capability of experienced AI engineers on realistic benchmarks. However, these agents have three weaknesses in efficiency that have not been fully addressed. First, only a small number of candidates can be executed within a realistic budget, so search rules that rank nodes by executed rewards, such as Monte Carlo-style tree search, rely on few and noisy scores and select the next node to explore less effectively. Second, no resource-aware strategy is used to schedule training jobs, which can lower hardware utilization and training efficiency. Third, every agent call is served by a single powerful model, which inflates inference cost. Here we introduce AIBuildAI-2.5, an agentic system that carries out the tree search with LLM agents and addresses each of the three issues. AIBuildAI-2.5 proposes a novel LLM-guided tree search, in which a judge scores each candidate on its expected improvement, grounding, and feasibility, and a selector ranks the pool of candidates from these scores and the state of the search. In addition, AIBuildAI-2.5 comprises a scheduler that launches training jobs with the current hardware resource status taken into account and a router that assigns lower-cost LLMs to less demanding tasks while reserving the most capable LLM for the most challenging sub-tasks in the AI model building workflow. AIBuildAI-2.5 ranks first on MLE-Bench with a medal rate of 73.3%, and outperforms a strong baseline on six autonomous AI research tasks from AIRS-Bench.
Read the full abstract

All authors Peijia Qin, Ruiyi Zhang, Qi Cao, Han Guo, Li Zhang, Pengtao Xie